Characterization of gas chemistry and noble-gas isotope ratios of inclusion fluids in magmatic-hydrothermal and magmatic-steam alunite

Chemical Geology 215 (2005) 155 – 184 www.elsevier.com/locate/chemgeo

Characterization of gas chemistry and noble-gas isotope ratios of inclusion fluids in magmatic-hydrothermal and magmatic-steam alunite G.P. Landis*, R.O. Rye U.S. Geological Survey, MS 963, P.O. Box 25046, Denver Federal Center, Denver, Colorado 80225, U.S.A. Accepted 1 June 2004

Abstract Chemical and isotope data were obtained for the active gas and noble gas of inclusion fluids in coarse-grained samples of magmatic-hydrothermal and magmatic-steam alunite from well-studied deposits (Marysvale, Utah; Tambo, Chile; Tapajo´s, Brazil; Cactus, California; Pierina, Peru), most of which are discussed in this Volume. Primary fluid inclusions in the alunite typically are less than 0.2 Am but range up to several micrometers. Analyses of the active-gas composition of these alunitehosted inclusion fluids released in vacuo by both crushing and heating indicate consistent differences in the compositions of magmatic-hydrothermal and magmatic-steam fluids. The compositions of fluids released by crushing were influenced by contributions from significant populations of secondary inclusions that trapped largely postdepositional hydrothermal fluids. Thermally released fluids gave the best representation of the fluids that formed primary alunite. The data are consistent with current models for the evolution of magmatic-hydrothermal and magmatic-steam fluids. Magmatic-steam fluids are vapordominant, average about 49 mol% H2O, and contain N2, H2, CH4, CO, Ar, He, HF, and HCl, with SO2 the dominant sulfur gas (average SO2/H2S=202). In contrast, magmatic-hydrothermal fluids are liquid-dominant, average about 88 mol% H2O, and contain N2, H2, CO2, and HF, with H2S about as abundant as SO2 (average SO2/H2S=0.7). The low SO2/H2S and N2/Ar ratios, and the near-absence of He in magmatic-hydrothermal fluids, are consistent with their derivation from degassed condensed magmatic fluids whose evolution from reduced-to-oxidized aqueous sulfur species was governed first by rock and then by fluid buffers. The high SO2/H2S and N2/Ar with significant concentrations of He in magmatic-steam fluids are consistent with derivation directly from a magma. None of the data supports the entrainment of atmospheric gases or mixing of air-saturated gases in meteoric water in either magmatic-hydrothermal or magmatic-steam fluids. Thus, the oxidation of SO2 to aqueous sulfate in the magmatic-steam fluids did not result from mixing with atmospheric oxygen. Both of the fluid types are characterized by high H2 contents that range from 0.2 mol% to the extraordinarily large amounts (66 mol%) observed in some magmatic-steam fluids. Modeling of gas speciation using SOLVGAS requires most of the gas species to have been in disequilibrium at the time of their trapping in the fluid inclusions. The origin of such extreme H2 concentrations, although problematic, is thought to be largely related to accumulation of H2 from the reaction of water with ferrous iron during the rise of

* Corresponding author. Tel.: +1 303 236 5406; fax: +1 303 236 6030. E-mail address: [email protected] (G.P. Landis). 0009-2541/$ - see front matter. Published by Elsevier B.V. doi:10.1016/j.chemgeo.2004.06.037

156

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

magma and probably even after exsolution of fluid from a magma. The large contents of reduced gases in the inclusion fluids are far in excess of those observed in volcanic emanations, and are thought to reflect the close bsampling positionQ of the host alunite relative to the location of the magma. Isotope ratios of He and Ne indicate largely crustal sources for these gases in the alunite parental fluids derived from Tertiary magmas, but a greater mantle component for the gases in alunite parental fluids derived from Proterozoic magmas. Published by Elsevier B.V. Keywords: Magmatic gas; Fluid inclusions; Gas chemistry; Hydrogen; Helium; Neon; Argon; Noble-gas isotopes; Alunite; Hydrothermal fluids

1. Introduction Alunite has proved to be a remarkably useful mineral in the study of magmatic, hydrothermal, and supergene processes, preserving isotopic and other geochemical evidence that characterizes the processes and environments of formation. The mineralogy, crystallography, thermodynamics, and geochronology of alunite-group minerals have been reviewed by Stoffregen et al. (2000). Stable-isotope characteristics and environments of formation of alunite have been reviewed by Rye et al. (1992), and the trace-element chemistry is reviewed by Deyell et al. (this Volume). The nature of fluid-inclusion and other gas sites, and the gas diffusion parameters and retention of primary gases in alunite fluid inclusions are described by Landis et al. (this Volume). Hydrothermal alunite occurs in three acid–sulfate environments: magmatic-hydrothermal, magmaticsteam, and steam-heated. Models showing the environment of deposition of these types of alunite and their relationship to the evolution of magmatic fluid are reviewed in several papers in this Volume. The quantitative aspects of phase relationships of magmatic fluids in the various environments have been reviewed by Fournier (1987, 1999), Giggenbach (1997), and Muntean and Einaudi (2001). Steamheated alunite is typically very fine grained, and suitably coarse samples were not available for the analyses reported here. Thus, only two of the hydrothermal environments of alunite formation are considered in this report. In the magmatic-hydrothermal environment, alunite is typically formed by the alteration of feldspar by H2SO4 derived from the disproportionation of magmatic SO2 in a condensing magmatic vapor plume that separated from a counterpart brine at the brittle– ductile transition at about 400 8C (cf. Smith and

Bruhn, 1984). The redox buffering of this vapor plume by its surrounding rocks near the brittle–ductile transition produces a fluid that is initially H2Sdominant. Both vapor and the residual saline fluids may be the source of later high-sulfidation ore deposition. This type of alunite most commonly replaces feldspars in igneous rocks and occurs in an assemblage of quartz+alunite+kaolinite+pyrite, with consecutive zones of alteration from vuggy-silica cores to peripheral propylitic+pyrite alteration. The y34S values of magmatic-hydrothermal alunite are large, and alunite–pyrite sulfur-isotope fractionations display equilibrium values that reflect the temperature of deposition. The magmatic-steam environment is characterized by veins of coarse-banded alunite thought to have precipitated from a rapidly rising and expanding magmatic vapor plume inferred to be SO2-dominant. The y34S values of the alunite are low (similar to the value for bulk sulfur in the magmatic vapor system) and clearly do not reflect equilibrium between reduced and oxidized sulfur species in the parental fluids. Presumably, the parental vapor rises so fast that the sulfur isotopic composition of the alunite reflects only that of the bulk sulfur in the system and not the values expected from the equilibration of reduced and oxidized aqueous sulfur species. It is currently unclear how the sulfate for magmatic-steam alunite is produced. The coarse-banded crystals may exhibit a sawtooth-shaped distribution of trace elements, and may have minor associated primary hematite, but have relatively little associated pyrite and rock alteration. This reconnaissance study presents the first quantitative chemical data on reactive- and noblegas isotope species contained in inclusion fluids in alunite from several well-documented localities. Fluid inclusions in magmatic-hydrothermal and magmatic-steam alunite were examined optically,

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

and data are presented on the gas compositions of inclusion fluids that were released by crushing and by subsequent heating in vacuo. There are similarities and differences of gas compositions between magmatic-hydrothermal and magmatic-steam alunite inclusion fluids, as well as among the fluids in each type of alunite from different localities. Neither type of alunite has fluid that contains atmospheric gases. The gas compositions are examined for approach to equilibrium among species using the computer program SOLVGAS (Symonds and Reed, 1993). Surprisingly, both the magmatic-steam and the magmatic-hydrothermal alunite show that the entrapped gases in the fluids were out of equilibrium largely because of the presence of H2 in amounts that far exceed those observed in volcanic gases. This H2 likely originates largely from the reaction of water with ferrous iron in the magma and is trapped in the alunite fluid inclusions because of the close proximity to the magmatic source of the fluids. In spite of these enigmatic high H2 contents, the results for magmatic-steam and magmatic-hydrothermal fluids from different localities are internally consistent and support existing models. Although this is a reconnaissance investigation and some of the discussion that it engenders is speculative, the consistency of the results suggests that further detailed studies on the reactive- and noble-gas isotope composition of alunite inclusion fluids are merited. Magmatic-steam and magmatic-hydrothermal alunite precipitate from aqueous fluids that can comprise both liquid and vapor. Both may contain gas or volatile components such as He, HCl, HF, SO2, H2S, and H2O that are released in a vacuum system for analytical measurements. In this paper, bgasQ or bvolatilesQ refer to species in the fluids rather than to the phase of the fluid, which can be supercritical or consist of one or two subcritical phases (liquid and vapor). Magmatic steam is an imprecise term that originates from the first description of magmatic-steam alunite (Rye et al., 1992) and implies a fluid that rose rapidly from a magma to shallow levels. Regardless of the phase relations in the fluid, magmatic steam refers here to a lowdensity fluid containing a variable component of water and may be dwetT or ddryT, depending on the degree of condensation of water vapor (Fournier, 1999).

157

2. Sample description We have focused our study on well-documented samples of magmatic-hydrothermal and magmaticsteam alunite that are available from previously published investigations. The samples of magmaticsteam alunite are from Marysvale, Utah (Cunningham et al., 1984; Rye et al., 1992), Cactus, California (Rye et al., 1992), Tambo, Chile (Deyell et al., this Volume), and Pierina, Peru (Fifarek and Rye, this Volume). The samples of magmatic-hydrothermal alunite are from Tapajo´s, Brazil (Juliani et al., this Volume), Tambo (Deyell et al., this Volume), and Pierina (Fifarek and Rye, this Volume). The two genetic types are distinguished by their geological occurrence, and by stable-isotope data that indicate a lack of sulfur isotopic exchange among aqueous sulfur species in their parental fluids for magmatic-steam alunite (Rye et al., 1992). Based on these criteria, the Marysvale alunite is clearly the archetype for the magmatic-steam environment. Where rates of fluid ascent were slower or distances from source magmas were greater, or where fluid–wallrock interaction was significant, various degrees of sulfur-isotope exchange between the aqueous sulfur species in the parental fluids may have occurred. These modifying effects on magmatic-steam fluids are best studied in the alunite at other areas, such as Tambo and Pierina. The designation of magmatic-steam and magmatichydrothermal alunite implies a genetic model for the processes that produced the parent fluids. We recognize that a continuum or broad overlap of processes and environments that produce these alunite types is likely. 2.1. Magmatic-steam alunite samples 2.1.1. Marysvale The best example of magmatic-steam alunite (Cunningham et al., 1984, 1996; Rye et al., 1992; Rowley et al., 1994; Rye, this Volume) occurs at Alunite Ridge in the Marysvale volcanic field in westcentral Utah above a 14 Ma intrusion. Coarse-grained alunite fills near-vertical fractures up to 20 m in width in intermediate-composition volcanic rocks overlying an unexposed, shallow intrusive body. The veins consist of multiple layers, each about 1–3 cm wide, of plumose alunite with minor hematite. Microprobe

158

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

studies have shown the presence of sawtooth patterns of P, Sr, and Ba across growth bands that are interpreted to represent burst and decay phenomena in the parent fluids (Cunningham et al., 1996; Rye, this Volume). The alunite is thought to have formed at a depth of b300 m from the paleosurface (Cunningham, personal communication) from fluids derived from direct degassing of a shallow (V1 km) epizonal magma; each alunite layer and sawtooth trace-element band represents a single degassing event or pulse of magmatic vapor. The y34S values of alunite are almost constant (~1x) and are the same as that of the bulk magmatic sulfur and sparse sulfides in the wallrocks (Rye et al., 1992). Observations of fluid inclusions indicate crystallization from a low-density fluid (Cunningham et al., 1984). 2.1.2. Cactus Magmatic-steam acid–sulfate alteration overlies gold and base-metal vein mineralization, of the dadulariaT–dsericiteT type, in the Middle Buttes Miocene quartz latite to rhyolite volcanic complex of the western Mojave Desert, California (Smith, 1941; Troxel and Morton, 1962). Alunite-producing alteration was roughly coeval with volcanism, and an Ar age of 18.4 Ma was obtained from fine-grained alunite (Bottaro, 1987). The magmatic-steam alunite occurs as coarse veins and volcanic breccia filling that cut steam-heated alteration without significant associated kaolinitic wallrock alteration. This alunite also has uniformly low y34S values, contains vapor-rich fluid inclusions, and does not appear to be related to the main Cactus mineralization. Low-temperature (130– 170 8C) intramineral oxygen isotopic data for this alunite suggest deposition in a shallow environment in which atmospheric oxygen might have been involved in the oxidation of sulfur gases to produce aqueous sulfate. The alunite intramineral oxygen-isotope thermometer has inherent uncertainties (Rye et al., 1992; Rye, this Volume), but if the Cactus sample is indeed of shallow low-temperature origin, the Cactus sample is an excellent one to test for possible involvement of atmospheric oxygen in the oxidation of sulfur gas species to aqueous sulfate. 2.1.3. Tambo Epithermal mineralization of the El Indio–Tambo deposits is hosted by intensely fractured rhyolitic to

dacitic, lithic ash-flow tuffs and other volcaniclastic rocks of Upper Oligocene to Upper Miocene age which are part of the El Indio–Pascua belt, Chile (Jannas et al., 1999; Deyell et al., this Volume). The main eruptive and hypabyssal volcanic activity occurred from 27 to 14 Ma, with isolated episodes of activity between 12 and 11.7 Ma. In the northern portion of the mineral belt, minor intrusions of dacitic dikes and eruptions of ignimbrite occurred up to 2 Ma. The deposits exhibit multiple stages of alunite formation representing both magmatic-hydrothermal and magmatic-steam acid–sulfate environments. Coarsely banded magmatic-steam alunite (Banded Stage alunite) occurs in veins up to 20-cm wide, and crosscuts both magmatic-hydrothermal alunite of pre-ore Stage 1 (11.0 Ma) and Stage 2 breccia alteration (8.7 Ma). Textural relations, geochronology, and stable-isotope data indicate that deposition of Stage 3 alunite (8.2 Ma) and magmatic-steam Banded Stage alunite overlaps in time. As will be discussed, the transitional nature of this alunite is supported by our gas data. The Tambo samples offer the opportunity to trace the evolution of fluids through different types and stages of alunite deposition. Stage 3 alunite is a transitional type between those of Stage 2 (magmatic-hydrothermal) and the Banded Stage alunite (magmatic-steam). 2.1.4. Pierina The Pierina Au–Ag deposit (Fifarek and Rye, this Volume) is in the Cordillera Negra, west of the Rio Santa and Cordillera Blanca batholith, within a largely volcanic section of calc–alkalic lavas, tuffs, and pyroclastic breccias, and minor epiclastic sedimentary rocks. The orebody occurs in rhyolite ash-flow tuffs overlying a footwall sequence of andesitic and dacitic flows. Penecontemporaneous with the rhyolite tuffs are breccias, tuffs, and flow domes in, and marginal to, a volcanic vent complex that is south of the orebody and represents small phreatic magmatic eruptions. Hydrothermal alunite in the orebody and sericitic alteration 4 km to the south near the domed vent seem to be contemporaneous and have been dated at about 14.5 Ma by K–Ar methods. Magmaticsteam alunite is rare. It occurs as moderately to coarsely plumose vein and breccia fillings (Stage 4) that postdate magmatic-hydrothermal alunite, but its temporal relationship to other stages of mineralization

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

is unknown. The magmatic-steam sample from Pierina exhibits a large increase in y34S (c8x) from that of the bulk sulfur composition (1x), implying rates of fluid flow sufficiently slow to permit significant sulfur-isotope exchange between reduced and oxidized aqueous sulfur species in the parent fluids (Fifarek and Rye, this Volume). 2.2. Magmatic-hydrothermal alunite samples 2.2.1. Tambo The Stage 2 magmatic-hydrothermal alunite of Deyell et al. (this Volume) is closely associated with early gold mineralization, but postdates major mineralized hydrothermal breccias. This alunite occurs as fine-grained (b50 Am) tabular crystals and as compositionally zoned overgrowths on bladed alunite that is up to 1 cm long on barite and breccia fragments. 2.2.2. Pierina The magmatic-hydrothermal alunite of Pierina formed during Stage 1 acid–sulfate alteration that overlaps and grades out from the central silicic alteration, forming part of the assemblage alunite– quartzFpyrophyllite, dickite, kaolinite. This alunite type occurs typically as pink, platy to tabular or bladed crystals up to 4 mm in length, commonly intergrown with pyrite. The rock alteration described by Fifarek and Rye (this Volume) indicates a progressive neutralization of highly acidic aqueous fluids that migrated outward from zones of more focused fluid flow. Fluid temperatures, based on alunite-pyrite sulfur isotopic data, were about 240 8C. 2.2.3. Tapajo´s Alunite, identified as magmatic-hydrothermal by Juliani et al. (this Volume), formed as part of an alteration assemblage related to high-sulfidation mineralization (1.87 Ga) genetically linked to the magmatic activity on what was then the margin of the Amazonian craton. Postcaldera volcanic rocks host the high-sulfidation mineralization that occurs in explosive hydrothermal breccia bodies and massive, hematite-rich, vuggy–silica caps. Quartz–alunite alteration, formed under a silica caprock, extends up to 200 m from the breccia centers and is known to occur up to 230 m below the surface. Of the five stages of alunite recognized at shallow levels (A1–A5), our

159

sample belongs to A5, a late-stage rose-colored, finegrained, randomly oriented alunite that crosscut and replaced the branching crystals of Stage A4 (Juliani et al., this Volume). Sulfur isotopic data for alunite and coexisting pyrite indicate formation at 130 to 400 8C, consistent with equilibrium between aqueous sulfide and sulfate in the parental fluids. The Tapajo´ s occurrence was remarkably little affected by later regional metamorphic or supergene events, largely because of burial in a shallow graben shortly after the mineralization event.

3. Fluid inclusions in magmatic-steam and magmatic-hydrothermal alunite Fluid-inclusion investigations, especially when combined with isotope data on host minerals, have long been one of the cornerstones of geochemical studies of ore-forming fluids in hydrothermal systems. Such studies have provided information on the temperature and salinity of fluids, as well as the chemistry and isotopic composition of components of both the ore-forming (primary and pseudosecondary inclusions) and the postdepositional (secondary inclusions) fluids (e.g., Roedder, 1984). Very little previous work exists on the fluid inclusions in alunite (cf. Cunningham et al., 1984). Fluid inclusions in alunite investigated in this study and in Landis et al. (this Volume) are generally V1 Am in diameter, have high gas/liquid ratios, and are too small for detailed temperature–salinity–clathrate studies (Roedder, 1984). However, temperature data often can be obtained from mineral assemblages or calculations of sulfur-isotope equilibrium fractionation for coexisting alunite and pyrite, and from oxygen-isotope fractionations between sulfate and the OH site in alunite (Rye et al., 1992). Photomicrographs in Fig. 1 illustrate rare 1–2 Am gas-rich inclusions in specimens of magmatic-steam alunite. Most inclusions in our samples are 0.2 to 0.01 Am and are gas-rich. Fluid inclusions in magmatichydrothermal alunite are described by Landis et al. (this Volume). Differences between fluid-inclusion populations in samples of magmatic-hydrothermal and magmaticsteam alunite were not observed. Both populations exhibited high gas/liquid ratios; and two-phase liquid–

160

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Fig. 1. Photomicrographs of fluid inclusions in magmatic-steam alunite. Width of the field of view of each image is 7.2 Am. (a) Cactus alunite with 0.9 Am vapor-dominant inclusion nucleated around 0.2 Am primary hematite grains. (b) Cactus alunite with unusually large 1.1 Am vaporrich inclusion. (c) Marysvale alunite vapor inclusions with faint wispy outlines of crystal domains made evident by the change in long-axis orientation of elongate inclusions. (d) Tambo alunite with 2.25 Am vapor inclusion near large primary hematite (dark) solid inclusions. (e) El Indio alunite with visible gas bubbles in a field otherwise predominantly of vapor-rich inclusions. Inclusion to left of X is 0.16 Am wide with a 0.09-Am-diameter gas bubble, and the inclusion below Y is similar in size. (f) El Indio alunite crystal domain exhibiting a high concentration density of vapor inclusions of 0.06 Am diameter. The El Indio alunite fluid-inclusion images are included because the deposit is similar to, and related to, the Tambo occurrence. See Landis et al. (this Volume) for magmatic-hydrothermal alunite fluid-inclusion images from the magmatichydrothermal deposits at Tapajo´s.

gas bubble inclusions only rarely were observed. The proportions of rare planes of large secondary fluid inclusions, along with ’dark’ semiopaque regions, were similar in both types of alunite. Although no

fluid inclusions were discernable in these regions, it is suspected that minor, but important, amounts of gas were released from these regions during crushing and thermal decrepitation of the samples.

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

161

Table 1 Data for fluid-inclusion gas; results for active gas by QMS Sample

y34Salunite y34SA yD Wt. N2

O2

Ar

H2

H2O H2S SO2

HCl HF

CH4 CO2 CO

He

Avg

ppm

ppm

%

%

ppm ppm

%

ppm

Magmatic-steam alunite Thermally released Marysvale M939a2 0.8 M939a3 M939a4 M341a 1.2 M341b Cactus Cactus 2
1.0 Tambo Stage 3 T02b 1.8 Tambo- banded alunite T08a 1.1 Pierina P452 8.5 Crush-Released Marysvale M939a1 M341 M341 M341 Cactus Cactus A Cactus 1 Tambo Stage 3 T02b Tambo-banded alunite T08a KB-02a Early 0.4 KB-02a Late KB-02b Early KB-02b Late Pierina P452

Avg Avg g

%

Avg 13.8 1459.0

ppm ppm

%

%

7722.8 21.3 48.7

3.9 807.3 157.0 1226.8 0.7

3.1 13.3

148 7997 1890 1507 164

1358 5460 2321 2181 1800

0 0 0 0 8

138 4088 1686 1153 76

6 561 435 160 142

0 292 558 3527 1075

1.80 1.97 0.95 0.83 0.27

0.56 1.29 0.79 0.94 8.18

93.98 5.67 0.00 3.50 1.28

251 732 1232 330 81

0.0


59 8.6 3.62 175

270

8.49 83.49 11

43

16

1282

0.08 3.04 1.10

24

2.0


78 6.6 0.52

59

654

6.68 91.13 12

30

41

2234

0.08 0.76 0.52

0

2.0


48 7.4 20.21 1045

54,992

48

3

673

5.0


85 9.5 1.98 146 Avg 21.2 1186.8

1.0

1.0


85 7.4 5.4 29.4
82 15.7 14.0

4.3 14.0 14.0 14.0

0.00 52.44 25.44 17.37 2.44

Magmatic-hydrothermal alunite Thermally Released Tambo Stage 2 T01a 27.3 2.0 Pierina P241 21.1 5.0 Tapajos 16/18 (200 8C-1) 26.7 5.0 16/18 (200 8C-2) Crush-released Tambo T01a Pierina P241 Tapajos 16/18


60

2.43 18.43 10.09 53.29 80.63

45.38 29.92

1

0.50 3.26

44

469 19.81 68.54 3 4 48 1400 0.17 9.30 0.00 40 4734.0 18.0 46.3 13.3 450.5 107.8 685.9 0.7 1.1 18.2 232.3

0.00 3189 36.58 545 33.72 116 20.51 77

1835 7385 8201 5360

1.73 35.81 12.26 15.32

15.62 10.74 38.54 53.74

0 0 0 0

3114 0 0 0

83 11 53 45

901 65 623 1124

4.37 0.02 0.12 0.20

5.3 36.85 2134 15.4 21.25 4464

4607 2667

48.67 10.35 16.29 55.70

0 0

1956

606 153

399 2971

1.78 1.27 0.00 1110 0.88 0.63 3.93 293

19

4431

65.65 4.04 12

24

9

90

7.4 48.62 127 0.59 0.91 1.40 2.40

14,228 298 534 785 715

13.59 0.17 0.41 0.42 0.41

21.72 98.80 98.00 97.30 96.70

22 22 26 14 77

43 52 81 79 56

10

0

0.14 0.07 0.14 0.18 0.14

9.5 53.24

10,496

23.18 0.26

0

0

0

0

0.16 0.67 21.44

6.6 19.54

2.0

1.05 18.29 61.92 23.25 6.86

303.7

Avg 0.6

11

74.74 97 15.30 6 12.77 457 7.68 649

0.43 2.17 7.71

1.14 20.13 0 0.27 70 0.37 125 0.39 120 0.31 92

2.3

0.0

0

275.9 8.8 87.9

44

200

1.44 97.55 17

3

8

2672

0.01 0.46 0.05

0


93 10.7 0.99 117

351

16.25 78.21 14

19

27

1913

0.09 4.21 0.00

0

0.20

11.1 17.6 2292.6 0.1

0

80.5


28 8.0

15.3

2.62 0.75 1.64 1.83

0.0


43 7.0 10.62 221 314,905 7.26 10.41 187 110 136 76 1.25 38.87 0.00 216 7.0 3.73 165 1321 10.44 58.43 292 94 117 0 3.89 22.77 0.51 212 Avg 13.5 63.5 3445.2 49.0 26.5 25.3 39.6 59.5 318.3 0.9 2.2 7.4 0.0 8.0 13.61

48

4070

65.09 11.30 20

38

47

313

0.21 1.64 7.69

0

10.7 13.31

79

2821

32.99 41.77 30

41

72

324

1.66 2.78 7.16

0

7.0 29.04

8

3017

1.51 0.86 34

8

6

30

0.41 0.16 67.71

11

162

Sample

Wt. g

Magmatic-steam alunite Thermally released Marysvale M939a2 M939a3 M939a4 M341a M341b Cactus Cactus 2 Tambo Stage 3 T02b Tambo-banded alunite T08a Pierina P452 Crush-released Marysvale M341 Cactus Cactus A Cactus 1 Tambo Stage 3 T02b

4

3

He

20

He


1

mol g

21

Ne


1

mol g

mol g


1

22

Ne

3

Ne


1

mol g

He/4He

R/Ra

20

Ne/22Ne

21

Ne/22Ne

4

He/21Ne

4

He/40Ara


1

mol g

Percent (%) magmatic Heb

0.15

8.98

0.08

818.46

0.039

1.58

7.904 9.067 9.297 9.554 3.698

0.113 0.031 0.045 0.029 0.346

1411.144 65.545 1426.165 3501.796 651.071

0.185 0.134 0.531 0.151 0.045

0.55 1.76 0.36 0.05 0.02

5.4 29.4 15.7 14.0

4.85E-13 2.21E-13 2.67E-13 5.73E-12

1.08E-19 1.48E-20 8.98E-21 1.70E-19

2.15E-12 3.23E-14 2.48E-14 9.39E-14

7.41E-15 1.55E-16 7.63E-17 8.80E-15

2.37E-13 3.47E-15 2.60E-15 2.54E-14

2.22E-07 6.70E-08 3.36E-08 2.97E-08

0.064 0.160 0.048 0.024 0.021

8.6

7.01E-13

3.16E-20

3.95E-11

1.24E-13

3.74E-12

4.51E-08

0.033

10.58

0.033

5.675

0.087

0.16

6.6

1.15E-12

1.86E-19

7.41E-12

2.04E-14

6.62E-13

1.62E-07

0.117

11.188

0.031

56.314

0.000

1.21

7.4

1.17E-13

4.33E-20

1.18E-12

5.15E-15

1.24E-13

3.71E-07

0.268

9.476

0.041

22.663

0.001

3.11

9.5

1.04E-12

7.09E-19

–

–

–

6.83E-07

0.494 0.57

8.21

0.19

49.26

0.084 0.049

5.93 6.93

14.0

6.02E-13

9.82E-20

1.05E-13

9.24E-15

2.99E-14

1.63E-07

0.118

3.512

0.309

65.077

0.056

1.23

5.3 15.4

2.67E-14 1.34E-14

2.24E-20 2.31E-20

1.97E-12 4.50E-14

7.23E-15 7.78E-17

2.27E-13 4.70E-15

8.39E-07 1.73E-06

0.606 1.249

8.683 9.591

0.032 0.017

3.699 171.916

0.241 0.110

7.35 15.40

6.6

3.60E-14

2.18E-20

4.55E-11

1.28E-13

4.18E-12

6.06E-07

0.438

10.884

0.031

0.281

0.000

5.24

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Table 2 Data for fluid-inclusion gas; noble gas by MAP

Tambo-banded alunite T08a Pierina P452

4.62E-13

3.26E-19

3.69E-13

1.90E-14

4.76E-14

7.07E-07

0.511

7.766

0.399

24.321

0.000

6.15

9.5

5.37E-13

3.83E-19

4.50E-13

1.78E-14

5.08E-14

7.14E-07

0.516

8.851

0.350

30.268

0.000

6.21

alunite 0.13

10.02

0.03

258.16

0.000

1.36

8.0

1.49E-12

1.51E-19

2.18E-12

6.87E-15

2.35E-13

1.01E-07

0.073

9.272

0.029

216.270

0.000

0.67

10.7

4.03E-13

1.03E-19

4.46E-13

1.34E-15

4.14E-14

2.55E-07

0.184

10.775

0.032

300.046

0.000

2.05

7.0 7.0 7.0

1.32E-13 3.32E-13 2.50E-15

1.93E-18 4.91E-18 3.61E-20

4.23E-14 1.40E-13 2.50E-16

1.35E-16 4.85E-16 7.30E-19

4.55E-15 1.61E-14 1.91E-17

1.47E-05 1.48E-05 1.44E-05

10.598 10.695 10.439 0.45

9.288 8.740 13.082 10.79

0.030 0.030 0.038 0.04

975.139 684.131 3418.042 65.88

0.001 0.160 0.000

52.95 53.43 52.15 5.44

8.0

8.61E-14

4.72E-20

4.76E-13

1.55E-15

4.11E-14

5.49E-07

0.397

11.574

0.038

55.479

0.000

4.72

10.7

5.03E-14

3.56E-20

2.04E-13

6.60E-16

2.04E-14

7.08E-07

0.511

10.003

0.032

76.283

0.000

6.16

7.0

1.81E-14

3.29E-20

2.53E-14

8.10E-17

2.72E-15

1.82E-06

1.313

9.305

0.030

223.679

0.004

6.47

All samples except Tapajo´s calculated on assumption of mixture of crustal and magmatic He Tapajo´s assumed to be a mixture of crustal and mantle/hot-spot He Ra is the 3He/4He of atmosphere=1.38410
6. a Approximate ratio from abundance of He and Ar concentrations of Table 1. b Calculated on assumption that crustal R/Ra=0.02, magmatic R/Ra=8.00, and mantle/hot-spot R/Ra=20.00.

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Magmatic-hydrothermal Thermally Released Tambo Stage 2 T01a Pierina P241 Tapajo´s 16/18 (200 8C-1) 16/18 (200 8C-2) 16/18 (200 8C-3) Crush-released Tambo Stage 2 T01a Pierina P241 Tapajo´s 16/18

7.4

163

164

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

All alunite samples exhibited three types of fluidinclusion occurrences. Type 1 primary inclusions are elongate parallel to the alunite c axis and are commonly concentrated at micrometer to submicrometer crystal-domain boundaries (Fig. 1a–c). These boundaries are defined by a change in contiguous optical properties and epitaxial growth of alunite in a subtly different crystal orientation in otherwise massive alunite. Type 2 primary inclusions are aligned parallel to surfaces of mineral deposition, such as along growth bands of plumose plates and tabular blades, and are elongate perpendicular to the surface of mineral deposition (Fig. 1e). Type 3 inclusions are randomly dispersed, somewhat elongate, without preferred orientation or location, but are always within the interiors of optically definable crystal domains (Fig. 1f). Planes of secondary fluid inclusions are extremely rare in coarse-grained magmatic-steam alunite. Daughter minerals, except hematite, were absent from the gas-rich inclusions, and could not be positively identified within rare, extremely small, liquid-rich inclusions. Primary hematite is common in alunite both as solid inclusions (Fig. 1d) and as nucleating centers for fluid inclusions (Fig. 1a).

4. Analytical methods The compositions of alunite inclusion fluids released both by crushing the sample and by heating in vacuo to 200 8C were determined. Crushing the sample presumably released gas from larger primary(?) and predominantly secondary fluid inclusions. Heating the sample presumably yielded gas predominantly from the submicrometer primary fluid inclusions. Whereas the crush-released inclusion fluids have a bpostdeposition of aluniteQ component contained in secondary inclusions, the thermally released inclusion fluids best represent the composition of the alunite parental fluid. The bpostdeposition of aluniteQ component, however, is thought to be

intimately related to the fluids responsible for the formation of alunite, and the component is not a younger or modern contaminant gas. The 1.87 Ga alunite from Tapajo´s, Brazil, that was examined here also was used by Landis et al. (this Volume) to define the He and Ar diffusion parameters in alunite, and to evaluate the long-term preservation of the composition of the fluids trapped in fluid inclusions in alunite. The Landis et al. study showed that the original gas contents of the fluids, including He and Ar and their isotopes, have remained trapped in the fluid inclusions for nearly 2.0 Ga. Likewise, the data for much younger (23 to 2 Ma) alunite examined in this study indicate retention of the initial gas compositions. A typical sample for analysis consisted of several ethanol-washed pieces of alunite, 0.5 to 1.0 cm on a side, totaling 4 to 30 g. The sample was crushed offline in a stainless steel tube sealed under vacuum, returned to the manifold, and the gases were extracted and analyzed. The sample then was heated in the crush tube to 200 8C for 60 min, and the evolved gases were analyzed. These extracted gases were analyzed both by a dynamically pumped quadrupole mass spectrometer (QMS: Pfeiffer Vacuum models Prisma and 410) for active-gas mixtures, and by a static high-resolution sector mass spectrometer (MAP: Mass Analyzer Products MAP 215-50) and QMS (in static mode) for noble-gas isotopic and elemental compositions. Analyzed concentrations of N2, O2, Ar, H2, H2O, H2S, SO2, HCl, HF, CH4, CO2, CO, and He were determined quantitatively by interpretation of QMS mass spectra and intensities. Mass spectra were corrected for background, specific ionization sensitivity relative to nitrogen, ion fragmentation, and for isobaric overlap of peaks with nominally identical atomic or molecular mass by a matrix solution to linear equations as reported by Landis and Hofstra (1991). The MAP analyses of He and Ne and their isotopes were performed separately after cryogenic separation on the same sample. Argon isotopic

Notes to Table 3: a Ratio temperatures calculated from thermodynamic relations in Giggenbach (1997): log [CO2/H2S]=0.0168T
9.83+3990/(T+273); log [CH4/CO2]=4625/(T+273)
10.4; and log [CH4/H2S]=(0.0168T)
20.23+(8615)/(T+273). b Correspondence temperature (CT) as defined in Symonds et al. (1994) and calculated with SOLVGAS (Symonds and Reed, 1993). T (8C) is approximate temperature of equilibrium between major gas species, defined when mole concentration measured and predicted from thermochemical calculations are within 10%. Hyphen indicates a range in temperature and a forward slash reports CT for two subgroups of gas species not in equilibrium with each other as discussed in the text.

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

165

Table 3 Gas ratios from Table 1 data Sample Magmatic-steam Thermally released Marysvale M939a2 M939a3 M939a4 M341a M341b Cactus Cactus 2 Tambo Stage 3 T02b Tambo-banded alunite T08a Pierina P452 Crush-released Marysvale M939a1 M341 M341 M341 Cactus Cactus A Cactus 1 Tambo Stage 3 T02b Tambo-banded alunite T08a KB-02a Early KB-02a Late KB-02b Early KB-02b Late Pierina P452 Magmatic-hydrothermal Thermally released Tambo Stage 2 T01a Pierina P241 Tapajos 16/18 (200 8C-1) 16/18 (200 8C-2) Crush-released Tambo T01a Pierina P241 Tapajos 16/18

log (CH4/CO2)

T (8C)a

526

0.51 0.18 0.08
0.05
1.48

151 164 168 174 245

2.52

927

325/1050

3.46

472


1.58

251

1.88

866

332–470

0.39

2.82

400


0.97

217

1.85

863

390

3.7

0.02

4.43

566


0.81

210

3.62

1024

670

42.2

0.89

4.46

568


1.75

262

2.71

944

0.0 49.5 41.1 38.3

0.22
1.50
1.13
0.97

162 247 226 218

80.0 79.7

0.15 0.15

166 166

N2/Ar

H2S/SO2

0.0 96.0 109.6 79.6 13.5

0.00 0.00 0.00 0.00 0.11

4.00

133.9

0.24

7.9

log (CO2/H2S)

T (8C)a

log (CH4/H2S)

T (8C)a

610 515/910 620

44.1

0.48

3.27

452


0.70

204

2.56

931

34.2 19.8 17.0 17.8 33.6

0.51 0.42 0.32 0.18 1.37

2.71 2.09 2.15 2.44 1.60

387 281 295 348 214


0.90
0.58
0.43
0.33
0.35

214 198 191 186 187

1.81 1.51 1.72 2.11 1.25

859 829 850 888 205


0.62

200

50.7

CT (8C)b

530–550

9.8

4.83

2.44

348


1.52

248

0.93

766

280/820

28.3

0.75

3.48

474


1.66

256

1.82

860

515

0.3 28.3

1.70 3.11

3.32 2.89

458 409


1.49
0.77

246 207

1.83 2.12

860 889

480/1160

33.4

0.53

2.90

410


0.89

213

2.02

879

47.2

0.74

2.96

418


0.22

182

2.74

947

96.3

4.25

1.66

214

0.43

154

2.09

886

166

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

analyses by MAP or QMS (40Ar/36Ar) were not possible in the early development stages of the Denver USGS noble-gas laboratory. With completed instrument development, data for Ar, Kr, and Xe will be included in future investigations. As discussed in Landis et al. (this Volume), the two-step crushing and heating extraction process released nearly all of the gas in the fluid inclusions. In most instances, crushing yielded only a small fraction (ca. 5–15%) of gas released subsequently by heating. Maintaining the temperature of alunite at 200 8C for 1 h released gas entrapped in abundant submicrometer primary inclusions. The high ratio of the volume of vapor/liquid (clow density) observed in the optical examination of fluid inclusions implies that gas from submicrometer inclusions was released largely by thermal expansion of alunite and less by fluid-inclusion decrepitation. The amount and nature of gas released from ’dark’ regions is unknown, but is presumably small, given their small volume percentage in alunite. The sustained heating at 200 8C did not release gas generated by thermal decomposition of alunite (Landis et al., this Volume). The compositions of gas from crush-released fluid potentially contain large amounts of extraneous gas components trapped, subsequent to alunite formation, along fractures and planes of secondary inclusions, whereas thermally released gas compositions most likely represent the volatiles that were originally trapped during alunite crystallization. Although these methods do not yield a clean separation of gases between primary and secondary inclusion fluids, the fluids are nonetheless predominantly one or the other. QMS analytical uncertainty is typically F0.5%, with calibration and other error propagation yielding a total estimated error of b2%. The QMS detection limit is b10 ppb v/V (parts per billion by volume or ppbV; also mole fraction) of total gas. Total procedural, calibration, and instrumental erros for MAP analyses are F0.5% to no more than F0.8% (1r as percent (%) of mean).

5. Results for fluid-inclusion gas chemistry Gas data from heating 16 and crushing 13 samples of magmatic-steam and magmatic-hydrothermal alunite from five localities are reported in Tables 1 and 2.

Analyses of He and Ne isotopes by MAP ion counting, as well as important calculated ratios, are also given (Tables 2 and 3). Data are grouped by inferred environment of alunite formation. A few generalizations are possible from Table 1. The order of gas abundance for magmatic-steam fluids is H 2 ONH 2 NN 2 NCOJCO 2 NArNCH 4 JO 2 NHFNSO 2 N HeNHClNH2S. The order of gas abundance for magmatic-hydrothermal fluids is H2OJH2NCO2J N2NHFJCH4NArNCONO2NHClNH2SNSO2NHe. Magmatic-steam fluids are the most volatile-rich and average about z20 times the N2, O2, and Ar, z2 times the H2, and about half the H2O content of magmatic-hydrothermal fluids. Magmatic-steam fluids are SO2-dominant, with SO2JH2S (average SO2/ H2Sc202), whereas magmatic-hydrothermal fluids have SO2VH2S (average SO2/H2Sc0.7). Concentrations of sulfur species of both fluids are low (tens to hundreds of ppm V), but are typical of the results obtained from studies of fluid-inclusion gases in other deposits (Landis and Rye, 1989; Goldfarb et al., 1989; Hofstra et al., 1991; Landis and Hofstra, 1991; Plumlee et al., 1994). Magmatic-steam fluids are enriched in HCl, but contain less HF relative to magmatic-hydrothermal fluids. Magmatic-hydrothermal fluids seem to have lost more He, which is highly mobile. It was possible to measure He routinely only in magmatic-steam fluids (to concentrations as low as 10 ppm V by QMS methods). The MAP, with about 106 more sensitivity, was unable to detect He in some magmatic-hydrothermal fluids. Some Marysvale magmatic-steam alunite inclusion fluids, with large He contents, have only SO2 as a sulfur gas. Methane, CO2, and especially CO are significantly elevated in magmatic-steam relative to magmatic-hydrothermal fluids. Both fluid types contain exceptionally high amounts of H2, in amounts never before detected in either fluid-inclusion gas of hydrothermal mineral deposits (Landis and Rye, 1989; Landis and Hofstra, 1991; Graney and Kesler, 1995) or in volcanic emanations (Symonds et al., 1994). The N2/Ar ratios of all fluids (Table 3) are not those of atmospheric or air-saturated waters (ASW), and He contents exceed the solubility limits in meteoric water by 1105 toN1108; the data preclude a significant component of atmospheric gases in either magmatic-steam or magmatic-hydrothermal fluids. As discussed below, most components of these gas mixtures are not in

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

thermodynamic equilibrium. Isotopic data for both He and Ne indicate a mixture of nonatmospheric radiogenic and nucleogenic 4He–21Ne–22Ne crustal components, and 3He–20Ne-enriched mantle sources (Table 2) in fluids for both magmatic-hydrothermal and magmatic-steam alunite.

167

evolved magmatic vapor fluids (magmatic-hydrothermal environment) that equilibrated with crystallized igneous rocks, formed large amounts of sulfuric acid from disproportionation of SO2, and ultimately mixed with meteoric water. 6.1. Thermally and crush-released inclusion fluids

6. Discussion Numerous authors have discussed models for the nature of magmatic-steam and magmatic-hydrothermal fluids (e.g., Rye et al., 1992; Rye, 1993; Rye, this Volume; Hedenquist and Lowenstern, 1994; Fournier, 1999). The data for gas offer the opportunity to compare the composition of the parental fluids from each genetic type of alunite and to infer the differences in their environments of deposition. Insofar as current models of fluid evolution are correct, the gas data offer an opportunity to make a comparison between the composition of low-density fluids (magmatic-steam environment) that were derived directly from a magma, and the composition of the condensed

6.1.1. Atmospheric gases in the hydrothermal fluids The data obtained by crushing presumably are for a bulk composition that includes the contributions of multiple generations of fluids trapped along secondary inclusion planes and crystal boundaries, together with a lesser compositional contribution from the fluids of submicron-sized primary inclusions. The gas contents of rare larger primary fluid inclusions opened by crushing are insignificant to the total amount of gas released in the crushing procedure. As demonstrated by the detailed studies of Tapajo´s alunite (Landis et al., this Volume), the composition of the gas in fluids released by crushing can differ from the composition released thermally. The crush-released gas shows no evidence that the fluids contain an atmospheric

Fig. 2. Plot of N2/Ar versus mole % N2 in fluids, showing values of atmospheric gases (=Air or Atm) and air-saturated water (ASW) at different temperatures. Solid symbols are compositions of thermally released fluids; open symbols are compositions of crush-released fluids. Magmaticsteam fluids=square, and magmatic-hydrothermal fluids=solid circle or triangle. Individual data-point labels used in this and subsequent plots are M=Marysvale, C=Cactus, P=Pierina, Ta=Tapajo´s, and T2=Tambo Stage 2 alunite, T3=Tambo Stage 3 alunite, and Tb=Tambo Banded alunite. The N2/Ar of atmosphere and the ASW values (dotted lines) increase from 38.4 (25 8C) to 44.9 (100 8C) to ~55 (200 8C). The effect of atmospheric gas entrainment and air-saturated water (e.g., meteoric water) incursion into the magmatic fluids would produce N2/Ar ratios between those for air and ASW, depending on the temperature of air saturation. For typical meteoric water—hydrothermal fluid conditions— almost all measured fluid compositions plot outside the range of those with possible Air–ASW contamination.

168

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

component, and it is believed that the differences in compositions of fluids released by heating and crushing reflect real differences in the nature of primary (alunite parental fluids) and later hydrothermal fluids as incorporated by primary and secondary inclusions, respectively. The data obtained from inclusion fluids extracted by heating best characterize the composition of the primary hydrothermal fluids responsible for the precipitation of the host alunite. As shown in Fig. 2 and Table 3, the absolute concentrations of N2 and Ar and the N2/Ar ratios of the fluid-inclusion gases indicate that neither atmosphere (Atm) nor air-saturated meteoric water (ASW) was a significant component of the fluids. Furthermore, the thermally released magmatic-steam and magmatic-hydrothermal inclusion fluids of alunite plot far from the compositions of atmosphere and air-saturated water at 25, 50, and 100 8C in the ternary N2–O2–Ar system (Fig. 3). Clearly, there is no

Fig. 3. Ternary plot of N2–O2–Ar; Atm, ASW, and symbols are defined as in Fig. 2. The three ASW compositions, which were calculated for 25, 50, and 100 8C, plot in the direction of increasing N2/O2 at higher temperature (Weiss, 1970, 1971). Only data for thermally released fluid are shown. The low oxygen content and the variable N2/Ar ratios preclude inclusion of atmospheric gas or dissolved ASW gas in either magmatic-steam or magmatic-hydrothermal fluids. Both fluid types show a broad overlap in composition fields, with magmatic-hydrothermal fluids having the lowest N2/Ar.

evidence of atmospheric gases in any of the inclusion fluids. 6.1.2. Comparison with volcanic gases Volcanic gases are predominantly mixtures of H2O+CO2+SO2, and the composition of alunite inclusion fluids can be compared with those for volcanic gas emanations (e.g., Rose et al., 1986; Symonds et al., 1994; Gerlach and McGee, 1994, 1998; Gerlach et al., 1999, 2002). The fluid-inclusion data for alunite are compared with the mixing line of X H2O–X CO2+SO2 in Fig. 4, wherein a fluid with a composition down and to left of the mixing line for volcanic gases must have significant amounts of other gas species, such as H2, CH4, CO, N2, or Ar. These species are not typical major components of volcanic gas emanations. The compositions of magmatic-steam fluids from Marysvale and Tambo are far removed from the mixing line in Fig. 4, and have a trajectory that suggests the presence of additional gases in the fluids. All data for magmatic-hydrothermal and the remaining magmatic-steam fluids plot sub-parallel to the mixing line, suggesting that other C-bearing species, N, or H are present. Data for the crush- and heat-released fluids plot within the same fields, implying that the secondary fluids are genetically linked to the primary fluids of alunite deposition. 6.1.3. N2–Ar–He–Ne The N2–Ar–He–Ne concentrations and ratios, and the He and Ne isotopic compositions of fluids, identify the sources of components and the volatile separation and condensation events in the fluids (Giggenbach, 1997; Moore et al., 2001). This section gives examples of how such data may be used to investigate the details of fluid evolution. Giggenbach’s N2–Ar–He ternary plot (Fig. 5) can be a tool to recognize boiling and condensation in fluids, and to record the partitioning of gases between liquid and vapor in magmatic-steam and magmatic-hydrothermal fluids (Norman and Musgrave, 1994; Giggenbach, 1997; Moore et al., 2001). In Fig. 5, magmatic-steam fluids plot in a broad field of elevated N2 and He contents, whereas magmatic-hydrothermal fluids plot along the N2–Ar join. For all gases analyzed, with the exception of those in the Tapajo´s fluids, the R/Ra (3He/4Heobserved/3He/4Heair) of He points to a major (90 to N98%) component of crustal He (R/Rac0.02,

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

169

Fig. 4. Plot of alunite inclusion fluid X H2O versus X CO2+SO2 as mole fractions. These three gases typically dominate volcanic gas compositions, and data from active volcanoes plot close to the mixing curve shown by the solid line (Symonds et al., 1994, and references therein). Fluid compositions from this study that plot down and to the left of the mixing line contain increasing amounts of other gas species, such as H2, CH4, CO, N2, Ar, and H2S. Data for thermally released gas (solid symbols) from magmatic-hydrothermal fluids (circles) plot systematically near this mixing line, as do the gas data of about half of the magmatic-steam fluids (squares). The compositions of some Marysvale and Tambo magmatic-steam fluids are well removed from the mixing line, indicating that the fluids contain major amounts of other gas species. Data for crush-released fluid inclusions from both magmatic-steam and magmatic-hydrothermal alunite mimic the compositional data for the thermally released magmatic-steam fluid, suggesting that most of the crush-released fluids were either from the same inclusion populations or from different (secondary) populations whose compositions were similar to those in the thermally released primary fluids. Data labels are sample numbers in Table 1.

low 3He/4He), mixed with a mantle-derived He component (R/Rac8, higher 3He/4He). Lack of oxygen in the fluids (Fig. 3), highly variable N2/Ar, and composition tie-lines that do not project back to atmosphere or air-saturated water compositions indicate that this crustal He was not derived at shallow levels by mixing with meteoric water in either magmatic-steam or magmatic-hydrothermal fluids. Whereas the fluids in Marysvale alunite, as characteristic of the end-member magmatic-steam environment, are rich in He, the magmatic-hydrothermal fluids have little or no He. As seen in the evolution of gas compositions of fluids from the Tambo alunite sequence of Stage 2 (magmatic-hydrothermal) through Stage 3 (magmatic-steam), to Banded alunite (magmatic-steam), even the late-stage magmatic-steam fluids from Tambo are depleted in He, a feature that suggests a different type of magmatic-steam environment than that at Marysvale. Whereas the He content of Marysvale magmaticsteam fluids is consistent with their derivation directly from a magma, the Tambo fluids show evidence of

degassing after separation from a melt, such as would be expected to occur in liquid-phase magmatichydrothermal fluids below the brittle–ductile transition. Thus, the magmatic-steam fluids at Tambo were likely derived from flashed liquids below the brittle– ductile transition, and not directly from a magma as at Marysvale. The plot of R/Ra for He versus the N2/Ar ratio of thermally and crush-released fluids in Fig. 6 further develops the concepts portrayed on the N2–Ar–He ternary diagram of Fig. 5 (Giggenbach, 1992). In the crush-released fluids, the abundances of secondary gases that have different compositions are greater than in the thermally released primary fluids. The tie-lines linking bpairedQ compositions for thermally and crush-released fluids from the same sample illustrate this point. The R/Ra of the He differentiates He derived from mantle (3He-enriched) versus crustal (strongly radiogenic, 4He-dominant) sources, whereas N2/Ar ratios reflect a combination of sources and the processes of liquid–vapor-phase separation. In hydrothermal fluids, the gas/liquid partition coefficients for

170

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Fig. 5. Ternary plot of N2–Ar–He. Data-point labels, Atm, ASW, and other symbols are as defined in Fig. 2. The different values for ASW that are calculated at 25, 50, and 100 8C plot towards Atm with increase in temperature. Almost no He is present in magmatichydrothermal fluids, indicating that they have degassed and lost He as the fluids recondensed. The variable He content in magmaticsteam fluids exceeds the solubility limits for ASW, such as might occur during initial exsolution of fluids from a magma. The field for magmatic-steam fluid represents a mixture of magmatic and crustal He, with gas chemistry modified by vapor-liquid phase fractionation. Thermally released data only are shown.

N2 and Ar differ by a factor of about 2, such that N2/ Ar of vapor will be approximately twice that of the liquid from which it separates (Weiss, 1970, 1971). Thus, condensed liquids will have lower N2/Ar ratios than their vapor counterpart. Thermally released fluids in Fig. 6 show a wide variation in N2/Ar, from b0.3 to 134, and have generally low R/Ra (average of 0.15), thus requiring a large component of crustal He. The compositions of all magmatic-steam thermally released fluids, especially those from Marysvale alunite (R/Rac0.05), plot near a crustal He-isotope value with a large range of N2/Ar. This range in N2/Ar indicates that magmatic-steam fluids contain both condensed liquids and vapor. The maximum possible N2/Ar ratio of vapor separated by distillation processes from condensed meteoric water is about 110, whereas the minimum N2/Ar of the liquid meteoric water residue is ~15 (Norman and Musgrave, 1994; Moore et al., 2001). The greater N2/Ar ratios of crushreleased fluids observed in most samples compared to

the ratios of the paired thermally released fluids most likely reflect late-stage, vapor-phase fluids trapped during the waning or post-crystallization stages of alunite deposition (cf. Norman and Musgrave, 1994; Giggenbach, 1997). The R/Ra variations in Fig. 6 reflect mixing of crustal and mantle He in the parental magma (Ozima and Podosek, 2002; Landis et al., this Volume). Compositions of magmatic-hydrothermal fluid plot entirely in the low N2/Ar ratio and higher R/Ra corner of Fig. 6, in contrast to the data for magmatic-steam gas. These data are consistent with prior gas/species partitioning between liquid and vapor phases that resulted in increased N2/Ar ratios in the vapor, and are consistent with the direct addition of mantle-derived He to the vapor phase in the fluids represented by the crush-released gases. The 4He/40Ar of these fluids (Table 2) can be approximated by the measurements of 4He and total Ar, assuming that Ar is dominantly 40Ar (295 atmc40Ar/36ArJ20,000 MORB; Ozima and Podosek, 2002). Almost all 4He and 40Ar are radiogenic, and the 4 He/40Ar ratio is constant for a given Th/U, K/U, and duration of accumulation. A typical constant crustal production ratio of 4He/40Ar is about 2 to 6 (maximum of ~102 to 103 for MORB; Ozima and Igarashi, 2000; Ozima and Podosek, 2002), whereas 4He/40Ar is approximately 0 to 0.5 in magmatic-steam and magmatic-hydrothermal fluids. The 4He/40Ar of magmatic-steam gas is greater than that in magmatichydrothermal gas, and for both genetic types of alunite, the R/Ra and 4He/40Ar ratios were larger for the crush- than the thermal-released gas. The content of mantle-derived He and Ar increased in the crushreleased fluids. Although the R/Ra indicates He is predominantly crustal in origin, the 4He/40Ar ratio is not crustal production. During magmatic vapor separation from its parent magma, z90% of the He was dlostT relative to Ar from the alunite-forming fluids. Neon in magmas can be a complex mixture derived from terrestrial, evolved upper mantle (MORB-like), deeper and more primitive mantle plumes (OIB-like), and subducted crustal–atmospheric (recycled) sources (Ozima and Podosek, 2002). The unique isotopic signatures (Kennedy et al., 1990; Farley and Poreda, 1993; O’Nions and Ballentine, 1993; Ballentine and Barford, 2000; Ozima and Igarashi, 2000) for these different sources can permit recognition of the magmas that gave rise to magmatic fluids. The data

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

171

Fig. 6. Plot of R/Ra versus N2/Ar, where R=3He/4He and Ra=R (atmosphere=1.38410
6). Helium R/Ra varies with mixing between crustal He having a radiogenic R/Ra (c0.02) and a mantle-magmatic He that is enriched in 3He with R/Rac8. The N2/Ar ratio varies with the source of magma, history of boiling, volatile separation, and condensation of fluids; the ratio is controlled by the relative solubilities of N2 and Ar in condensed fluids, and by the Bunsen coefficients for liquid–gas partitioning for He isotopes, N2, and Ar. The plotted data exhibit highly variable magmatic N2/Ar and He isotope ratios that reflect these mixing–condensation–boiling processes. Tie-lines connect thermally and crush-released fluid compositions. Atm and ASW and other symbols are defined in Fig. 2. The crush-released fluids cannot be derived from the mixing of fluids compositionally similar to thermally released fluids with ASW because the measured He contents of alunite inclusion fluids are 105 to more than 108 those in ASW (c5–44 Acc/kg H2O STP). Thus, gas compositions in crush-released fluid most likely represent volatiles in hydrothermal fluids trapped shortly after alunite deposition. The legend identifies symbols used for magmatic-steam and magmatichydrothermal fluids. Symbols with solid circle centers are the compositions of crush-released gas, and the open symbols are for thermal release. Crush- and thermal-released fluid pairs from the same sample are connected with dashed tie-lines. Atm and ASW (N2/Ar=38.4 at 25 8C=55.1 at 200 8C) compositions are plotted for reference. A small 3He/4He isotopic fractionation between air and water is neglected. Compositions of Tapajo´s thermally released fluid plot off scale at N2/Ar=0.33 and 28.26 and R/Ra=10.6 and 10.7 ratios, respectively.

are evaluated with a three-isotope plot of 20Ne/22Ne versus 21Ne/22Ne (Fig. 7). The Ne in magmatic-steam inclusion fluids (20Ne/22Ne=9.1–11.2) and magmatichydrothermal (20Ne/22Ne=8.7–13.1) alunite seems to be a mixture of a minor amount of nucleogenic crustal Ne in combination with a major amount of magmatic Ne that contains a MORB mantle component (20Ne/22Nec12.5) for the Tertiary fluids, and that contains an OIB mantle–plume component (20Ne/22Nez13) for the 1.87 Ga Tapajo´s fluids. A more primitive OIB mantle–plume Ne isotopic signature at Tapajo´s is compatible with the less-evolved Proterozoic continental margin geological setting described by Juliani et al. (this Volume). The high concentrations of radioactive U, Th, and K in the crust, and the associated alpha and neutron decay products, induce nucleogenic production of Ne isotopes. In crustal nucleogenic processes, an increase in

21

Ne/22Ne from magmatic Ne at 0.029 involves the nucleogenic production of 21Ne by 18O(a,n)21Ne and 24 Mg(n,a)21Ne, and a 20Ne/22Ne decrease from 9.82 by 19F(a,n)22Ne and 25Mg(n,a)22Ne. The paths of Mg production would be most important in rocks that have high Mg abundances, such as those in the lower crust and mantle. Both crustal nucleogenic processes are only weakly evident in the data in Fig. 7. None of the Ne isotopic data in Fig. 7 are far removed from magmatic or atmospheric values. Both 4 He and 21Ne should be produced in the upper crust by radiogenic decay and nucleogenic interactions at a nearly constant ratio (Kyser and Rison, 1982; Kennedy et al., 1990; O’Nions and Ballentine, 1993; Yatsevich and Honda, 1997). This production ratio, given typical upper crustal U–Th concentrations, is estimated to be 4He/21Nec2.1107. The maximum 4 He/21Ne in Table 2 (3.5103) is four orders of

172

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Fig. 7. Plot of 20Ne/22Ne versus 21Ne/22Ne. Main plot is of data only for thermally released (200 8C) fluid, and the insert shows at an expanded scale the data both for thermal and crush releases. Magmatic-atmospheric Ne, MORB (mid-oceanic ridge basalt; Honda et al., 1993), Yellowstone OIB (oceanic island basalt; Craig et al., 1978), and Loihi-Kilauea (Sarda et al., 1988) correlation lines, and the nucleogenic range of values are included for reference. The Ne is a mixture of both MORB and OIB with magmatic and crustal nucleogenic Ne.

magnitude less than crustal production, even given a low R/Ra for crustal He. The R/Ra of He appears to increase with 21Ne/22Ne ratios, yet incorporation of crustal He into the magmatic fluids does not seem to be accompanied by significant amounts of crustal Ne. Fluids of the three stages of Tambo alunite exhibit a general loss of He and a progressive increase in R/Ra (i.e., more mantle-derived 3He) and a decrease in 4 He/21Ne in the progression from magmatic-hydrothermal to magmatic-steam environments. The combined N2–Ar–He–Ne data for magmatichydrothermal and magmatic-steam fluids suggest that: (1) alunite in these environments formed in the presence of two-phase fluids; (2) magmatic-steam fluids contained crustal He in amounts far larger than those derivable from ASW; (3) Ne was derived largely from magmatic and mantle sources, with only minor contributions from nucleogenic crustal sources; and (4) magmatic-hydrothermal fluids lost almost all of their He as a result of boiling and vapor-phase removal. Although these fluids exsolved at relatively shallow depths in the crust, their rare-gas isotopic compositions originated in sources located in lower crust and mantle rocks.

6.2. Active-gas data Fluids from magmatic-steam and magmatic-hydrothermal alunite both contain similar amounts of CO2. The CO2 content varies inversely with H2O in magmatic-hydrothermal fluids, whereas CO2 varies positively with H2O in magmatic-steam fluids (Table 1). These relationships are consistent with the low CO2/H2O expected for magmatic-hydrothermal fluids, wherein CO2 content is controlled by its solubility in condensing fluids (e.g., magmatic gas scrubbing); in contrast, high CO2/H2O would be expected for magmatic-steam fluids that are released directly from a magma (Doukas and Gerlach, 1995; Zimbelman et al., 2000; Symonds et al., 2001). A plot showing the speciation of carbon gas among CO2–CO–CH4 (Fig. 8) indicates that all magmatic-hydrothermal fluids are CO2-dominant, whereas only some magmatic-steam fluids are CO2-dominant. Marysvale magmatic-steam fluids in alunite show the greatest variation in speciation, with significant amounts of CO and CH4 in some samples. For the Marysvale fluids in Fig. 8, a tie-line projected from the CO apex to the CO2–CH4 binary

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

173

magmatic-hydrothermal fluids (cf. Giggenbach, 1997). 6.3. Redox equilibrium and excess H in alunite inclusion fluids

Fig. 8. Plot of CO2–CO–CH4 relative abundances of magmatichydrothermal and magmatic-steam fluids. All magmatic-hydrothermal fluids are CO2-dominant. Magmatic-steam fluids contain more CO and CH4. Data for Marysvale alunite inclusion fluids suggest a constant CH4/CO2 ratio with an apparent equilibrium temperature of 168 8C (Giggenbach, 1997). Localities are indicated by letters as in Fig. 2.

defines a CH4/CO2 ratio that equates to a calculated thermodynamic temperature of 168 8C (Giggenbach, 1997). Similar calculated temperatures based upon CO2/H2S, CH4/CO2, and CH4/H2S gas ratios are tabulated in Table 3. These temperatures vary from 150–262 8C for CH4/CO2 of all fluids, 400–568 8C for CO2/H2S, and 860–1025 8C for CH4/H2S. The different calculated thermodynamic temperature ranges for each gas/species ratio indicate systematic disequilibrium gas speciation in the parent fluids. This disequilibrium is further evident in the discussion of SOLVGAS calculations in Section 6.4. The SO2–HCl–H2S plot (Fig. 9) illustrates the SO2-dominant nature of magmatic-steam fluids over a range in HCl concentrations. Magmatic-hydrothermal fluids are consistently H2S-rich relative to magmaticsteam fluids. This preferential enrichment of reduced and oxidized sulfur gas/species in magmatic-hydrothermal and magmatic-steam fluids, respectively, is also shown in the SO2–CO2–H2S plot in Fig. 10. Fluids from both environments contain similar amounts of CO2. These observations require different redox states ( fO2 conditions) for magmatic-steam and

The H2–SO2–He plot in Fig. 11 illustrates that both magmatic-steam and magmatic-hydrothermal fluids are H2-rich, and that Marysvale magmatic-steam and Tapajos magmatic-hydrothermal fluids in alunite differ in their SO2-to-He contents compared to each other and to fluids from other localities. These differences are consistent with different fluid histories at Marysvale, Tapajo´s and Pierina, Tambo, and Cactus as further discussed in Section 6.7. It is well known that H2 will partition into the vapor phase during and after exsolution of fluids from a magma (Symonds and Reed, 1993). The alunite inclusion fluids are significantly more H2-rich (0.2 to 66 mol%, F1 mol%) than can be inferred from thermodynamic and log fO2 relations, discussed below, or than has been reported for volcanic gas

Fig. 9. Plot of SO2–HCl–H2S relative abundances in magmatichydrothermal and magmatic-steam fluids. Sulfur dioxide-dominant magmatic-steam fluids and H2S-dominant magmatic-hydrothermal fluids have different relative HCl contents. Magmatic-hydrothermal fluids exhibit a restricted range of relative HCl concentrations whereas the relative HCl concentrations of magmatic-steam fluids are highly variable and reach values approximately nine times those observed in magmatic-hydrothermal fluids. Localities are indicated by letters as in Fig. 2.

174

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

parental fluid at the growth surface of the precipitating mineral. That is, on a scale of several micrometers, large chemical gradients form in the fluid at the fluid– crystal interface, and the composition of the fluids trapped in very small fluid inclusions therefore differs from that of the parental fluid. Disequilibrium gas compositions can therefore occur in very small inclusions. The results for synthetic fluid inclusions in quartz suggest that, if the precipitation of alunite involved reactions that released H2, e.g., KCl þ 3AlCl3 þ 2H2 SO4 þ 6H2 O ¼ KAl3 ðSO4 Þ2 ðOHÞ6 þ 5Cl2 þ 5H2

Fig. 10. Plot of SO2–CO2–H2S relative abundances in magmatichydrothermal and magmatic-steam fluids, both of which contain similar relative CO2 concentrations but are readily distinguished by their different H2S/SO2 ratios. Localities indicated by letters as in Fig. 2.

emissions (0.5 to 3.5 mol%). Although high H2 contents are recognized in gases of active magmas (Gerlach and Nordlie, 1975; Gerlach, 1980, 1993b; Symonds and Reed, 1993), the contents are not nearly as high as observed in some of our samples. The H2 contents in fluid inclusions may be the result of: (a) postdepositional reactions in the fluid inclusions; (b) selective trapping during the formation of very small fluid inclusions; (c) reaction of water with ferrous iron in the magma (2FeO+H2O=Fe2O3+H2); (d) dissociation of water (e.g., H2O=H2+1/2 O2) or other reactions (e.g., SO2+2H2O=H2SO4+H2) during and (or) after separation of a fluid from the magma; or (e) the addition of H2 that separated from deeper magmatic fluids. The fluid inclusions lack the possible reactionproduct gas species necessary to generate the measured H2 contents. Furthermore, it is not likely that the high H2 content of alunite inclusion fluids resulted from the selective trapping in small fluid inclusions. The mechanics of gas trapping have been examined in individual synthetic fluid inclusions in quartz (Landis, unpublished data). In these studies, rapidly sealed (quenched) fluids in small inclusions (b5 Am diameter) commonly are not in equilibrium with the

ð1Þ

then H2 could be trapped in disequilibrium amounts in submicrometer fluid inclusions during depositional growth of alunite. Alunite source fluids appear generally to be of low density but of highly variable water content. Disequilibrium trapping of H2 by this reaction would be more effective in a high-density aqueous fluid, and if this or similar reactions were the dominant source of H2, a positive correlation between H2O and H2 in fluids released from small inclusions would be expected (not stoichiometric). Fig. 12 shows

Fig. 11. Plot of H2–SO2–He relative abundances in magmatic-steam and magmatic-hydrothermal fluids. Magmatic-steam fluids are strongly characterized by the abundance of these three gas species. The magmatic-steam end-member fluids from Marysvale alunite are the most enriched in SO2 and He, whereas H2 is generally equally abundant both in magmatic-steam and magmatic-hydrothermal environments.

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

175

volatiles at the surface by traditional methods. The underlying magmatic fluids may have contributed to further enrichment of the H2 contents of alunite inclusion fluids, as proposed in Section 6.7. 6.4. Equilibrium speciation and fO2 of magmaticsteam and magmatic-hydrothermal fluids

Fig. 12. Plot of H2O–H2 contents in mole fraction of magmatichydrothermal and magmatic-steam fluids. Hydrogen concentrations increase with decreasing water content, indicating that H2 abundances are not an artifact of compositions attained at the fluidinclusion interface, but were actually present in the fluid as it was trapped (see text). Both crush- and thermal-release fluid compositions are plotted. Magmatic-steam fluids (n); magmatic-hydrothermal fluids (o). The X H2O=X H2 mixing line is shown for reference, and X H2 is plotted in log units to expand the scale.

X H2O versus X H2 of the alunite fluids, with an H2O-H2 mixing line for reference. The H2O content of the fluids exhibits an inverse linear correlation with that of H2. The importance of chloride and fluoride complexes to vapor transport of chemical components necessary to alunite formation from a low-density fluid is unknown, but H2 contents do not correlate with that of HCl or HF derived from important ligands such as KCl or (AlF3)2–AlF2O (cf. Symonds et al., 1992). The high H2 contents in alunite fluid inclusions are therefore concluded to reflect predominantly those of the primary hydrothermal fluids rather than depositional interface effects. The high H2 contents of magmatic fluids were likely created initially from the breakdown of H2O in the magma as it reacted with ferrous iron (Sato, 1978), and the creation of H2 probably started while the magma was at great depths and may have continued during the rise of these fluids to higher levels even after they left the magma. Hydrogen is the most mobile of magmatic gases, and it is not hard to imagine that it would escape from a magma at a greater rate than water and other volatiles (cf. Sato and Wright, 1966). The high H2 contents are recognizable in the alunite inclusion fluids because they have bsampledQ the magmatic volatiles much closer to their magma source than is possible by sampling volcanic

The data for the gas of thermally and crushreleased magmatic-steam and magmatic-hydrothermal alunite inclusion fluids are representative of the composition of the parental fluids, with possible modifications by retrograde reactions of gas species since their entrapment within the fluid inclusions. The SOLVGAS computer program and thermochemical database (Symonds and Reed, 1993) were used to compute gas equilibria by solving simultaneously a set of mass-balance and mass-action equations written for all gas species in the alunite fluid-inclusion data for each sample. The thermodynamic equilibrium gasspeciation calculations over a range of temperature (100–1200 8C) and pressure (100 to 1105 kPa) did not predict the elevated H2 contents observed in our samples and did not predict the observed SO2, CO, and CH4 concentrations. The data in Table 1 were recast into thermochemical components as input to SOLVGAS. A search was made for convergence of correspondence temperatures, or CT, for the gas mixture of each sample (Symonds et al., 1994). CT is the temperature (at a selected pressure) at which the logarithm of calculated/measured mole fraction of a species equals zero (i.e., log (X Calc/X Meas)c0, or X Calc/X Measc1). Equilibrium convergence is determined when the CTs of all considered gas species mutually approach zero. A practical limit of F10% equilibrium CT divergence (Symonds et al., 1994) was used; failure to converge implies a disequilibrium gas mixture. Multiple CTconvergence partial equilibrium conditions were achieved with the calculations, and efforts to correct measured input compositions for hypothetical causes of disequilibrium (Gerlach and Casadevall, 1986; Symonds et al., 1994) did not improve the partial equilibrium results. The alunite fluid-inclusion data do not represent equilibrium gas mixtures, but rather partial equilibrium (cf. Barton, 1963); that is, some gas species appear to approach equilibrium as indicated by agreement between calculated and

176

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

measured gas/species concentration at the same CT, but others do not. The log (X Calc/X Meas) results of SOLVGAS analyses are plotted against temperature in Fig. 13. Deposition of most magmatic-steam and magmaticmatic-hydrothermal alunite is thought to occur at less than 1 km or 10 mPa (e.g., Hedenquist and Lowenstern, 1994; Bethke et al., this Volume). Because fluid pressure does not have a large effect on the calculated results and is only important for gas species coupled by reactions with large DV, the effect of pressure on the calculations is not presented here. Equilibrium gas speciation changes dramatically with temperature. Approximate CTs, and a partial gas/ species high- and low-temperature CT, are illustrated

in the plots of log (X Calc/X Meas) versus temperature, and the indicated convergent-temperature estimates are listed in Table 3. This analysis implies equilibrium among gas species at multiple temperatures and, therefore, disequilibrium in the speciation of the total gas. This result of partial equilibrium is not unexpected for magmatic-steam alunite inclusion fluids, which are considered to have undergone rapid ascent from a magma, but the result is surprising for magmatichydrothermal fluids, considered to have equilibrated with rocks below the brittle–ductile transition. By not applying constraints of gas and rock buffers, which specify a log fO2 in the SOLVGAS calculations, the log fO2 at P–T–X can be determined from the calculated equilibrium of the component gas/

Fig. 13. Plot of example results from SOLVGAS thermodynamic equilibrium calculations. Equilibrium gas speciation was calculated from the starting alunite fluid-inclusion gas compositions in Table 1, at various temperatures. A Correspondence Temperature (CT) is obtained at the point where the log (Xi Calc/Xi Meas) is within approximately 10% for each major gas species. Xi is the calculated and measured mole fractions of species diT. Evaluation of measured gas compositions indicates disequilibrium to partial thermodynamic equilibrium, and at a log fO2 that is well below predicted conditions based upon mineral assemblages (hematite+alunite) and log fO2 buffers (NNO–MH–QFM). (a) Magmatichydrothermal Tambo T01a: partial equilibrium at approximately 300 8C for H2–H2O–CO2–CH4, and at 820 8C for CO2–CH4–SO2–H2S. (b) Magmatic-hydrothermal Tapajo´s1: poor convergence of CTs at about 490 8C. (c) Magmatic-steam Marysvale M341b: partial CT equilibrium at 410 and at 1050 8C. (d) Magmatic-steam Marysvale M939a4: CT convergence equilibrium temperature of 618 8C.

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

species input for each calculation. The calculated log fO2 versus temperature for each thermally released gas is plotted in Fig. 14. Also plotted is the estimated log fO2 (Luhr and Melson, 1996; Evans and Scaillet, 1997) of the anhydrite-bearing dacitic magma that erupted in 1991 from Mount Pinatubo, Philippines (NNO+1.7, at 780 8C and 220 mPa), and released about 18106 tonnes of SO2. The sulfur–gas redox speciation of magmatic-steam and magmatic-hydrothermal fluids reflects the fundamentally different nature of fluid processes in the two environments. Fluids released from deep I-type magmas have oxidation states along the QFM and NNO buffers, resulting in H2S-rich to SO2-rich fluids (Gerlach, 1986, 1993b; Giggenbach, 1987; Hedenquist, 1995). However, fluid reactions with wallrocks below the brittle–ductile transition typically result in H2S-rich fluids at higher levels in the crust (Ohmoto

177

and Rye, 1979; Rye, this Volume). Magmatic fluids (such as those at Marysvale) released from shallow high-temperature magmas are more likely to be SO2rich (Symonds et al., 1990, 1994; Snyder et al., 2001). Even deep magmas can have SO2-rich fluids because increasing the fO2 from QFM to QFM+2(=logfO2 of QFM+2 log units) also increases, from ~5% to N80%, the proportion of sulfur dissolved as sulfate in the melt (Carroll and Rutherford, 1988; Carroll and Webster, 1994). Sulfur dioxide becomes the dominant sulfur species in the fluid at logfO2 zQFM-1 (O’Neill, 1987), such as observed for the 1991 eruptions of Mount Pinatubo. The H2S/SO42
of higher level fluids is controlled by equilibrium with rocks below the brittle– ductile transition at about 400 8C in the magmatichydrothermal environment. The H2S/SO42
of magmatic-hydrothermal fluids may move away from control by rock buffers above the brittle–ductile

Fig. 14. Plot of log fO2 versus temperature, showing the oxygen fugacity predicted from SOLVGAS calculations based on gas compositions obtained in this study as described in Fig. 13. HM=hematit–magnetite buffer (calculated from Chase, 1998), NNO=nickel–nickel oxide buffer (Luhr and Melson, 1996; Evans and Scaillet, 1997; Huebner and Sato, 1970), SO2–H2S isomolar univariant (calculated from SOLVGAS and NIST-JANAF data; Symonds and Reed, 1993; Chase, 1998), and QFM=quartz–fayalite–magnetite buffer (NIST-JANAF, Chase, 1998; O’Neill, 1987). All log fO2 curves fall at or below the QFM fO2 buffer over the entire range of temperature at 100 kPa. The analyzed gas compositions recast to thermodynamic equilibrium are not compatible with observed mineral assemblages or other volcanic gas data. At 100 mPa, the curve for Marysvale M939a4b fluid crosses the HM buffer into the hematite-stable region. In this calculation, the composition was adjusted by taking the total H2 content of the fluids and recasting it to 0.5% H2, 3.5% SO2 and remainder H2O (solid line), and to 0.5% H2 with remainder SO2 (dashed line). Marysvale fluid composition M939a3 adjusted to (a) 100 kPa, (b) 100 mPa, and (c) 100 kPa and H2 recalculated to 0.5% with the remainder as H2O. The lowest log fO2 curve is for Marysvale M939a2, which has a very unusual content of 94% CO. The abundance of reduced-species H2 and CO depresses the Marysvale fluid log fO2 curves (Table 1). The fluids in the Tambo sequence of three samples correlate with an increase in H2 from 1.44% (magmatic-hydrothermal: Stage 2) to 6.68% (magmatic-steam: Stage 3) to 45.38% (magmatic-steam: Banded alunite). Tambo fluids show an increase in H2 and SO2, with a decrease in H2S with successive alunite deposition events. Pierina log fO2 magmatic-steam and magmatic-hydrothermal curves are not significantly different.

178

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

transition, during the period when SO2 disproportionates and HSO4
is formed in a condensing vapor plume, and when the Fe in the rock is removed by reaction with H2S to produce pyrite. The calculated log fO2 for all magmatic-steam and magmatic-hydrothermal fluids as a function of temperature (Fig. 14) ranges from the QFM buffer (Tambo Stage 2) to about QFM-7.5 (Marysvale M939a2). Marysvale magmatic-steam fluids are among the most reducing, and the log fO2 values of other magmatic-steam fluids completely overlap those for magmatic-hydrothermal fluids. Marysvale fluids yield a calculated log fO2 of about
24 to
27.2 at the approximate 610 8C CT, and Pierina magmaticsteam and magmatic-hydrothermal fluids yield a log fO2 of about
24.8. At 200 to 350 8C, the log fO2 of all alunite inclusion fluids (
41blog fO2b
33) are below the hematite+alunite stablility field (Crerar and Barnes, 1976; Stoffregen, 1987; Deyell, 2002). The calculated log fO2 values of both magmatic-steam and magmatic-hydrothermal alunite inclusion fluids are much too low for parental fluids, as indicated by the mineral assemblage alunite+hematite. 6.5. Summary of redox conditions of magmatic-steam and magmatic-hydrothermal fluids Of all of the magmatic-steam inclusion fluids analyzed, those from Marysvale seem to be the most closely linked directly to magma. Compared to Marysvale, the magmatic-steam fluids from Pierina and Tambo have higher H2O and H2S, and lower H2 contents, that are consistent with less-rapid ascent of fluids after exsolution from the magma. Slower rates of fluid ascent allow for greater diffusive separation of H2, greater condensation of H 2O, greater H2S production from the disproportionation of SO2, and more isotopic equilibration among aqueous sulfur species in the fluids (Rye, this Volume; Fifarek and Rye, this Volume). In contrast, magmatic-hydrothermal alunite inclusion fluids that are derived from below the brittle– ductile transition at about 400 8C are initially H2Sdominant (Symonds and Reed, 1993; Rye, 1993) because the ratio of their reduced to oxidized sulfur species is buffered by wallrocks. Later, at higher levels, these fluids may become SO42
-dominant as SO2 in a condensing vapor plume disproportionates to

produce sulfuric acid and overwhelms the rock buffer. The amount of H2 observed in the alunite magmatichydrothermal inclusion fluids is also in excess of the equilibrium fO2 values controlled by either rock or fluid buffers. Below about 400 8C, SO 2 in a condensing magmatic vapor fluid disproportionates to H2SO4 and H2S (Holland, 1965; Rye et al., 1992): 4SO2 þ 4H2 O ¼ 3H2 SO4 þ H2 S

ð2Þ

The H2S reacts with Fe in the rocks to produce pyrite, and the H2SO4 leads to the formation of alunite. The high H2 content observed in the inclusion fluids suggests that reduction of SO2 may also occur in the magmatic-hydrothermal environment by (i) reaction with the disequilibrium amounts of H2 in the fluids SO2 þ 3H2 ¼ H2 S þ 2H2 O

ð3Þ

or (ii) by coupled reaction with Fe in the host rock SO2 þ 2H2 þ 2Fe3 O4 ¼ H2 S þ H2 O þ 3Fe2 O3

ð4Þ

Possible reaction (4), however, must be of limited extent because portions of the host rock in many highsulfidation hydrothermal alunite occurrences are so thoroughly altered that nothing remains but vuggy silica. The low fO2 observed in both the magmaticsteam and magmatic-hydrothermal alunite inclusion fluids clearly needs further investigation. 6.6. Gas solubility in magmas, and mantle and crustal gas sources The incorporation of gas species into a fluid that separates from an ascending magma will be accompanied by chemical partitioning and fractionation as described by the gas-specific Henry’s Law constants (reviewed by Carroll and Holloway, 1994). Giggenbach (1996, 1997) has shown that a fluid phase will start to separate from a cooling and ascending magma at depths of ~60 km, near that of arc magma generation. At this depth, nearly 90% of the H2O will remain in the melt, but the H2O content will decrease with crystallization and declining pressure until most is released by the time the melt is near 1 km of the surface. However, N90% of the He, Ne, and CO2 will be exsolved from the melt into the fluid phase by the time the magma reaches depths of 20 km. The solubility of noble gases in silicate melts has received extensive attention (e.g., Carroll and Web-

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

ster, 1994; Shibata et al., 1998; Ozima and Podosek, 2002). It can be concluded from these investigations on molecular gas solubility in magmas that He and Ne in the alunite inclusion fluids must have exsolved into a supercritical phase at great depths while most of the H2O remained in the ascending parental magma. Until reaching shallow depths in the upper crust where fluids become subcritical and phase separations occur as fluids cross the brittle–ductle transition into fractured rock, both magma and separated fluid remain intimately intermixed. The fluid remains an open system, with input of fluids both exsolved from the magma and from surrounding crust. The 1.87 Ga Tapajo´ s magmatic-hydrothermal fluids in alunite contain He with an R/Ra of 19.5, which is much higher than the ratio in the younger magmatic-steam and magmatic-hydrothermal fluids (0.02 to 0.27). None of these fluids contains significant Atm or ASW gases, and our data suggest that the He in the magmatic-steam and magmatic-hydrothermal fluids is largely crustal-derived, but the Ne is largely mantle-derived. This apparent contradiction of crustal and mantle sources for these gases is easily explained by the model developed by Fournier (1999) wherein supercritical fluids exsolve at depth and ascend as a separate phase with their parental magma. The low R/Ra of ~0.02 for He indicates that it and other magmatic volatiles migrated in a separate vapor phase that entrained crustal He from the surrounding rocks as the vapor phase ascended through the crust. Helium, because of its greater diffusive mobility, is released selectively from radiogenic and nucleogenic sites within the crust and is readily incorporated into the supercritical fluid coexisting with rising magma, largely masking any original mantle He-isotope signature in the magma. Rock permeabilities in the mid- to lower-crust readily support fluid-solute transport over large rock volumes (Manning and Ingebritsen, 1999). However, Ne is more tightly bound in minerals in the crust and, consequently, is not added in sufficient quantities to mask the original mantle Neisotope signature of the magma. 6.7. Magmatic-steam and magmatic-hydrothermal model for alunite Significant differences as well as similarities are apparent in the gas chemistry of magmatic-steam and

179

magmatic-hydrothermal alunite inclusion fluids. More research is needed to understand better the importance of the high H2 contents in magmatic-steam and magmatic-hydrothermal fluids, and the importance of disequilibrium gas speciation in which the required calculated log fO2 values of the parental fluids are not compatible with the stability of the host alunite and associated hematite. A conceptual chemical model of magmatic-steam and magmatic-hydrothermal fluid evolution is illustrated in Fig. 15. The calculated log fO2-temperature relations from SOLVGAS are summarized in a band that is plotted as individual curves in Fig. 14. The starting point is composition A which represents the compositions of magmatic volatiles at approximately the conditions of partial equilibrium identified by SOLVGAS, but at realistic SO2/H2S ratios near the isomolar (SO2/H2S=1) curve as calculated by Ohmoto (1986). Magmatic-steam fluids of composition A in Fig. 15 rapidly rise from a magma into open fractures above the brittle–ductile transition along a trajectory that crosses the magnetite–hematite buffer into the stability field labeled dMS AluniteT. This trajectory is the most likely one for the parental fluids of Marysvale magmatic-steam alunite. Alternatively, composition A fluids could also rise slowly so that they react with wallrocks to composition B, and then rapidly ascend across the brittle–ductile transition along a different trajectory into the field of magmatic-steam alunite. This would be a possible path for the parental fluids of the Pierina magmatic-steam alunite. Magmatic-hydrothermal alunite parental fluids equilibrate with crustal rocks below the brittle–ductile transition prior to their ascent to shallow levels (Rye, this Volume; Bethke et al., this Volume). As vapor-phase fluids rise from the brittle–ductile transition and condense in open fractures, where SO2 disproportionates to form SO42
and leads to the formation of magmatic-hydrothermal alunite, the residual liquids below the brittle–ductile transition lose most of their volatiles, especially He, and undergo a decrease in N2/Ar. Such a liquid-phase fluid might be positioned at composition C. Volatiles such as H2 from the underlying magma may continue to be added to the magmatic-hydrothermal liquid fluids, or a pulse of H2-rich fluids from the magma may mix with magmatic-hydrothermal vapor-phase fluids at higher levels. The resulting fluids would be H2S- and CO2-dominant, high in H2 devoid of He, and

180

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Fig. 15. Plot of log fO2 versus temperature, illustrating an evolution model for deposition of magmatic-steam and magmatic-hydrothermal alunite that is consistent with the gas chemistry of alunite fluid inclusions. The fO2 buffers are as defined in Fig. 14, and calculated log fO2 of magmatic-steam and magmatic-hydrothermal alunite inclusion fluids are defined within the labeled gray region. Data for the 1991 dacite eruption of Mount Pinatubo (Luhr and Melson, 1996; Evans and Scaillet, 1997) are plotted to indicate strongly SO2-dominant magmatic gas conditions. Vertical rectangle is diagrammatic and shows the approximate temperature of the brittle–ductile transition. The circles show the modeled composition of magmatic-hydrothermal and magmatic-steam fluids based on the data in this study. The presumed starting fO2 of magmatic-steam fluids immediately after exsolution from a magma is shown by A; B shows the presumed fO2 of evolved magmatic fluids after ascent to just below the brittle–ductile transition and equilibration with crystalline rocks. Compositions A and B are representative of possible magmatic-steam fluids if pressure release is rapid. C is a possible composition of magmatic-hydrothermal fluids that may receive contributions of reduced gases from underlying fluids during the rise of a vapor phase above the brittle–ductile transition. The starting conditions of magmatic gas were approximated from magmatic-steam and magmatic-hydrothermal fluid partial equilibrium SOLVGAS-calculated correspondence temperatures and fO2, and from other constraints derived from geological and mineralogical conditions and gas data.

would yield the low calculated (disequilibrium) log fO2 observed for the magmatic-hydrothermal fluids.

7. Conclusions This study is the first attempt to analyze the gas components of the small fluid inclusions in crystals of magmatic-hydrothermal and magmatic-steam alunite. Although this investigation is a reconnaissance, and some of the discussion that it engenders is speculative, the consistency of the results for magmatic-hydrothermal and magmatic-steam alunite inclusion fluids is remarkable. The results not only support existing models for alunite formation (Rye, this Volume) but also are consistent with the geological and isotopic variations observed for magmatic-hydrothermal and

magmatic-steam alunite at different deposits (cf. Fifarek and Rye, this Volume; Deyell et al., this Volume; Bethke et al., this Volume). Such consistency supports the conclusions of Landis et al. (this Volume) that alunite-hosted fluid inclusions retain their primary fluid compositions. These compositions can provide important clues both to the environments and mechanisms of alunite formation and to the general chemical composition and source(s) of magmatic volatiles. Further detailed studies of alunite inclusion fluids in individual deposits are merited. The discovery of large, disequilibrium amounts of H2 in all of the alunite inclusion fluids is problematic and requires further study. However, it is important to note that the analyzed alunite samples have incorporated magmatic gases much closer to their sources than has been possible from any surface sampling of

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

magmatic gases. Generalized conclusions that can be made are: (1)

Other than for large disequilibrium amounts of reduced gases, alunite fluid inclusions preserve magmatic volatile compositions that are broadly similar to those derived from thermodynamic equilibrium speciation models (Symonds and Reed, 1993) based on reported compositions for volcanic gases (Gerlach and Nordlie, 1975; Giggenbach and Le Guern, 1976; Gerlach, 1993a; Giggenbach, 1997). (2) The relative abundance of gas species for magmatic-steam fluids is H2ONH2NN2NCOJ CO 2 NArNCH 4 JO 2 NHFNSO 2 NHeNHClNH 2 S, and for magmatic-hydrothermal fluids the relative abundance is H 2OJH 2NCO 2JN2NHFJ CH4NArNCONO2NHClNH2SNSO2. Magmaticsteam alunite forms from low-density SO2-dominant fluids, whereas magmatic-hydrothermal alunite forms from higher density H2S-dominant fluids. (3) The data show no evidence that atmospheric gas or air-saturated water (meteoric water) was involved in either the magmatic-steam or magmatic-hydrothermal fluids. Oxidation of SO2 to aqueous sulfate in magmatic-steam fluids was not accomplished by entrained atmospheric oxygen. (4) Sulfur dioxide dominates magmatic-steam fluids; concentrations are as high as 4088 ppmV and have an average SO2/H2S of 202. Magmatichydrothermal fluids are more H2S-rich, with an average SO2/H2S of 0.7. These results are consistent with the derivation of magmatic-steam fluids directly from a magma at shallow depth, and with magmatic-hydrothermal fluid derivation largely from condensed evolved magmatic fluids that accumulated near the top of the magma chamber below the brittle–ductile transition. (5) Multiple partial equilibrium reactions among different gas species (e.g., H2O–H2, SO2–H2S, CO2–CH4) at different P–T conditions are preserved in the gas mixtures. The data for neither the magmatic-hydrothermal nor the magmaticsteam alunite fluid-inclusion gas are representative of equilibrium gas mixtures. These fluids were in disequilibrium when they were trapped in the inclusions. These results are not surprising for

(6)

(7)

181

magmatic-steam fluids, given their presumed rapid ascent, but somewhat similar results for magmatic-hydrothermal fluids are surprising. The extraordinary high concentrations of H2 gas that characterize both magmatic-steam and magmatic-hydrothermal alunite inclusion fluids are enigmatic. The H2 is believed to be largely derived from the reaction of water with ferrous iron in the magma. The accumulation of H2 in disequilibrium amounts in the hydrothermal fluids is believed to be related to its high mobility relative to other components of magmatic fluids. The H2 contents in magmatic-steam alunite parental fluids are probably derived directly from a magma. The high H2 contents of the magmatic-hydrothermal alunite parental fluids may have resulted from the dfeedingT of additional magmatic volatiles to the low-density magmatic-hydrothermal fluids at higher levels. The He isotopic data indicate a crustal source for He, whereas the Ne isotopic data indicate a large mantle component for Ne in the alunite inclusion fluids. This difference suggests that these and probably other magmatic gas species separated as a fluid phase from the silicate melt at considerable depth in the crust, and that the fluid phase incorporated additional gases from crustal sources as it ascended with the melt.

Acknowledgements We thank Cari Deyell, Rich Fifarek, Skip Cunningham, and Caetano Juliani for providing the welldocumented samples for this work. Discussions with Mark Reed and Phil Bethke were extremely valuable, as were reviews by Paul Barton, Phil Bethke, and Bob Seal. [PD]

References Ballentine, C.J., Barford, D.N., 2000. The origin of air-like noble gas in MORB and OIB. Earth Planet. Sci. Lett. 180, 39 – 48. Barton Jr., P.B., 1963. Equilibrium in ore deposits. Spec. Pap.Miner. Soc. Am. 1, 171 – 185. Bethke, P.M., Rye, R.O., Stoffregen, R.E., Vikre, P., 2005. Evolution of the magmatic-hydrothermal acid-sulfate system

182

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

at Summitville, Colorado: integration of geological, stableisotope, and fluid-inclusion evidences. Chem. Geol. 215, 281– 315 (this volume). Bottaro, J.L., 1987. Geology of the Middle Buttes volcanic complex, Mojave district, Kern County, California. MSc thesis, San Jose State Univ., California. Carroll, M.R., Holloway, J.R. (Eds.), 1994. Volatiles in Magma, Rev. Mineral., vol. 30. Carroll, M.R., Rutherford, M.J., 1988. Sulfur speciation in hydrous experimental glasses of varying oxidation state: results from measured wavelength shifts of sulfur X-rays. Am. Mineral. 73, 845 – 849. Carroll, M.R., Webster, J.D., 1994. Solubilities of sulfur, noble gases, nitrogen, chlorine, and fluorine in magmas. In: Carroll, M.R., Holloway, J.R. (Eds.), Volatiles in Magma, Rev. Mineral, vol. 30, pp. 231 – 279. Chase Jr., M.W., 1998. NIST-JANAF thermochemical tables, 4th Edition. J. Phys. Chem. Ref. Data 9. 1951 p. Craig, H., Lupton, J.E., Welham, J.A., Poreda, R., 1978. Helium isotope ratios in Yellowstone and Lassen Park volcanic gases. Geophys. Res. Lett. 5, 897. Crerar, D.A., Barnes, H.L., 1976. Ore solution chemistry: V. Solubilities of chalcopyrite and chalcocite assemblages in hydrothermal solution at 200 to 350 8C. Econ. Geol. 71, 772 – 794. Cunningham, C.G., Rye, R.O., Steven, T.A., Mehnert, H.H., 1984. Origins and exploration significance of replacement and veintype alunite deposits in the Marysvale volcanic field, west central Utah. Econ. Geol. 79, 50 – 71. Cunningham, C.G., Rye, R.O., Bethke, P.M., Logan, M.A.V., 1996. Formation of coarse-grained vein alunite by degassing of an epizonal stock. Chapman Conference on Crater Lakes, Terrestrial Degassing and Hyper-Acid Fluids in the Environment. Extended Abs., Sept. 4–9, 1996, Crater Lake, OR. American Geophysical Union, Washington, DC, p. 27. Deyell, C.L., 2002. Synchronous Au–alunite in the Pascua-Lama high sulfidation system: constraints on the origin and chemistry of mineralizing fluids. Geol. Soc. Am. Abstr. Programs 34 (6), 186. Deyell, C.L., Rye, R.O., Landis, G.P., Bissig, T., 2005. Alunite and the role of fluids in the Tambo high-sulfidation deposit, El Indio–Pascua belt, Chile. Chem. Geol. 215, 185–218 (this volume). Doukas, M.P., Gerlach, T.M., 1995. Sulfur dioxide scrubbing during the 1992 eruptions of Crater Peak, Mount Spurr Volcano, Alaska. In: Keith, T.E.C. (Ed.), The 1992 Eruptions of Crater Peak Vent, Mount Spurr Volcano, Alaska, U.S. Geol. Surv. Bull., vol. 2139, pp. 47 – 57. Evans, B.W., Scaillet, B., 1997. The redox state of Pinatubo dacite and the ilmenite–hematite solvus. Am. Mineral. 82, 625 – 629. Farley, K.A., Poreda, R.J., 1993. Mantle neon and atmospheric contamination. Earth Planet. Sci. Lett. 114, 325 – 339. Fifarek, R.H., Rye, R.O., 2005. Stable-isotope geochemistry of the Pierina high-sulfidation Au–Ag deposit, Peru: influence of hydrodynamics on SO2
4 H2S isotopic exchange in magmaticsteam and steam-heated environments. Chem. Geol. 215, 253– 279 (this volume).

Fournier, R.O., 1987. Conceptual models of brine evolution in magmatic-hydrothermal systems. U.S. Geol. Surv. Prof. Pap. 1350 (2), 1487 – 1506. Fournier, R.O., 1999. Hydrothermal processes related to movement of fluid from plastic into brittle rock in the magmatic-epithermal environment. Econ. Geol. 94, 1193 – 1211. Gerlach, T.M., 1980. Investigation of volcanic gas analyses and magma outgassing from Erta’Ale lava lake, Afar, Ethiopia. J. Volcanol. Geotherm. Res. 7, 415 – 441. Gerlach, T.E., 1986. Exsolution of H2O, CO2, and S during eruptive episodes at Kilauea Volcano, Hawaii. J. Geophys. Res. 91, 12177 – 12185. Gerlach, T.E., 1993a. Thermodynamic evaluation and restoration of volcanic gas analyses: an example based on modern collection and analytical methods. Geochem. J. 27, 305 – 322. Gerlach, T.E., 1993b. Oxygen buffering of Kilauea volcanic gases and the oxygen fugacity of Kilauea basalt. Geochim. Cosmochim. Acta 57, 795 – 814. Gerlach, T.E., Casadevall, T.S., 1986. Evaluation of gas data from high-temperature fumaroles at Mount St. Helens, 1980–1982. J. Volcanol. Geotherm. Res. 28, 107 – 140. Gerlach, T.M., McGee, K.A., 1994. Total sulfur dioxide emissions and pre-eruption vapor-saturated magma at Mount St. Helens, 1980–88. Geophys. Res. Lett. 21, 2833 – 2836. Gerlach, T.M., McGee, K.A., 1998. Rates of volcanic CO2 degassing from airborne determinations of SO2 emission rates and plume CO2/SO2: test study at Pu’u’O’o cone, Kilauea volcano, Hawaii. Geophys. Res. Lett. 25, 2675 – 2678. Gerlach, T.E., Nordlie, B.E., 1975. The C–O–H–S gaseous system: Part II. Temperature, atomic composition, and molecular equilibria in volcanic gases. Am. J. Sci. 275, 377 – 394. Gerlach, T.M., Doukas, M.P., McGee, K.A., Kessler, R., 1999. Airborne detection of diffuse carbon dioxide emissions at Mammoth Mountain, California. Geophys. Res. Lett. 26, 3661 – 3664. Gerlach, T.M., McGee, K.A., Elias, T., Sutton, A.J., Doukas, M.P., 2002. Carbon dioxide emission rate of Kilauea volcano: implications for primary magma and summit reservoir. J. Geophys. Res. 107, 2189 – 2203. Giggenbach, W.F., 1987. Redox processes governing the chemistry of fumarolic gas discharges from White Island, New Zealand. Appl. Geochem. 2, 143 – 161. Giggenbach, W.F., 1992. Magma degassing and mineral deposition in hydrothermal systems along convergent plate boundaries. Econ. Geol. 87, 1927 – 1944. Giggenbach, W.F., 1996. Chemical composition of volcanic gases. In: Scarpa, R., Tilling, R.I. (Eds.), Monitoring and Mitigation of Volcano Hazards. Springer Verlag, Berlin, pp. 221 – 256. Giggenbach, W.F., 1997. The origin and evolution of fluids in magmatic-hydrothermal systems. In: Barnes, H.L. (Ed.), Geochemistry of Hydrothermal Ore Deposits, 3rd ed. Wiley, New York, pp. 737 – 796. Giggenbach, W.F., Le Guern, F., 1976. The chemistry of magmatic gases from Erta’Ale, Ethiopia. Geochim. Cosmochim. Acta 40, 25 – 30. Goldfarb, R.J., Leach, D.L., Rose, S.C., Landis, G.P., 1989. Fluid inclusion geochemistry of gold-bearing quartz veins of the

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184 Juneau Gold Belt, southeastern Alaska—implications for ore genesis. In: Keays, R.R., Ramsay, R.H., Groves, D.I. (Eds.), The Geology of Gold Deposits: The Perspective in 1988, Econ. Geol. Monogr., vol. 6, pp. 363 – 375. Graney, J.R., Kesler, S.E., 1995. Gas composition of inclusion fluid in ore deposits: is there a relation to magmas? In: Thompson, J.F.H. (Ed.), Magmas, Fluids, and Ore Deposits, Mineral. Assoc. Can. Short Course, vol. 23, pp. 221 – 245. Hedenquist, J.W., 1995. The ascent of magmatic fluid: discharge versus mineralization. In: Thompson, J.F.H. (Ed.), Magmas, Fluids, and Ore Deposits, Mineral. Assoc. Can. Short Course, vol. 23, pp. 263 – 289. Hedenquist, J.W., Lowenstern, J.B., 1994. The role of magmas in the formation of hydrothermal ore deposits. Nature 370, 519 – 527. Hofstra, A.H., Leventhal, J.S., Northrop, H.R., Landis, G.P., Rye, R.O., Birak, D.J., Dahl, A.R., 1991. Genesis of sediment-hosted disseminated gold deposits by fluid mixing and sulfidization: chemical reaction–path modeling of ore depositional processes documented in the Jerritt Canyon district, Nevada. Geology 19, 36 – 40. Holland, H.D., 1965. Some applications of thermochemical data to problems of ore deposits: II. Mineral assemblages and the composition of ore-forming fluids. Econ. Geol. 60, 1101 – 1166. Honda, M., McDougall, I., Patterson, D.B., 1993. Solar noble gases in the Earth: the systematics of helium–neon isotopes in mantle derived samples. Lithos 30, 257 – 265. Huebner, J.S., Sato, M., 1970. The oxygen fugacity–temperature relationships of manganese oxide and nickel oxide buffers. Am. Mineral. 55, 934 – 952. Jannas, R.R., Bowers, T.S., Petersen, U., Beane, R.E., 1999. Highsulfidation deposit types in the El Indio District, Chile. In: Skinner, B.J. (Ed.), Geology and Ore Deposits of the Central Andes, Soc. Econ. Geol. Spec. Pub., vol. 7, pp. 27 – 59. Juliani, C., Rye, R.O., Nunes, C.M.D., Snee, L.W., Corre´a Silva, R.H., Monteiro, L.V.S., Bettencourt, J.S., Neumann R., Neto, A.A., 2005. Paleoproterozoic high-sulfidation mineralization in the Tapajo´s gold province, Amazonian craton, Brazil: geology, mineralogy, alunite argon-age, and stable-isotope constraints. Chem. Geol. 215, 95–125 (this volume). Kennedy, B.M., Hiyagon, H., Reynolds, J.H., 1990. Crustal neon: a striking uniformity. Earth Planet. Sci. Lett. 98, 277 – 286. Kyser, T.K., Rison, W., 1982. Systematics of rare gas isotopes in basic lavas and ultramafic xenoliths. J. Geophys. Res. 87, 5611 – 5630. Landis, G.P., Hofstra, A.H., 1991. Fluid inclusion gas chemistry as a potential minerals exploration tool: case studies from Creede, CO, Jerritt Canyon, NV, Coeur d’Alene district, ID and MT, southern Alaska mesothermal veins, and mid-continent MVT’s. J. Geochem. Explor. 42, 25 – 59. Landis, G.P., Rye, R.O., 1989. Reconnaissance gas chemistry of the Creede, Colorado, hydrothermal system. U.S. Geol. Surv. OpenFile Rep. 89-0084. Landis, G.P., Snee, L.W., Juliani, C., 2005. Evaluation of alunite argon ages and fluid-inclusion integrity: stepwise noble-gas heating experiments on 1.87 Ga alunite from Tapajo´s Province, Brazil. Chem. Geol. 215, 127–153 (this volume).

183

Luhr, J.F., Melson, W.G., 1996. Mineral and glass compositions in June 15, 1991, pumices: evidence for dynamic disequilibrium in the dacite of Mount Pinatubo. In: Newhall, C.G., Punongbayan, R.S. (Eds.), Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines. PHIVOLS and Univ. Washington Press, pp. 733–749. Manning, C.E., Ingebritsen, S.E., 1999. Permeability of the continental crust: implications of geothermal data and metamorphic systems. Rev. Geophys. 37, 127 – 150. Moore, J.N., Norman, D.I., Kennedy, B.M., 2001. Fluid inclusion gas compositions from an active magmatic-hydrothermal system: a case study of The Geysers geothermal field, USA. Chem. Geol. 173, 3 – 30. Muntean, J.L., Einaudi, M.T., 2001. Porphyry–epithermal transition: Maricunga belt, northern Chile. Econ. Geol. 96, 743 – 772. Norman, D.I., Musgrave, J.A., 1994. N2–Ar–He compositions in fluid inclusions: indicators of fluid source. Geochim. Cosmochim. Acta 58, 1119 – 1131. Ohmoto, H., 1986. Stable isotope geochemistry of ore deposits. In: Valley, J.E., Taylor Jr., H.P., O’Neil, J.R. (Eds.), High Temperature Geological Processes, Rev. Mineral., vol. 16, pp. 491 – 560. Ohmoto, H., Rye, R.O., 1979. Isotopes of sulfur and carbon. In: Barnes, H.L. (Ed.), Geochemistry of Hydrothermal Ore Deposits, 2nd ed. Wiley, New York, pp. 509 – 567. O’Neill, H.St.C., 1987. Quartz–fayalite–iron and quartz–fayalite– magnetite equilibria and the free energy of formation of fayalite (Fe2SiO4) and magnetite (Fe3O4). Am. Mineral. 72, 67 – 75. O’Nions, R.K., Ballentine, C.J., 1993. Rare gas studies of basin scale fluid movement. Philos. Trans. R. Soc. London 344, 141 – 156. Ozima, M., Igarashi, G., 2000. The primordial noble gases in the earth: a key constraint on Earth evolution models. Earth Planet. Sci. Lett. 176, 219 – 232. Ozima, M., Podosek, F.A., 2002. Noble Gas Geochemistry. 2nd ed. Cambridge Univ. Press, New York. Plumlee, G.S., Leach, D.L., Hofstra, A.H., Landis, G.P., Rowan, E.L., Viets, J.G., 1994. Chemical reaction path modeling of ore deposition in Mississippi Valley-type Pb–Zn deposits of the Ozark region, U.S. midcontinent. Econ. Geol. 89, 1361 – 1383. Roedder, E., 1984. Fluid inclusions. Rev. Mineral. 12, 644 p. Rose, W.I., Chuan, R.L., Giggenbach, W.F., Kyle, P.R., Symonds, R.B., 1986. Rates of sulfur dioxide and particle emissions from White Island volcano, New Zealand, and an estimate of the total flux of major gaseous species. Bull. Volcanol. 48, 181 – 188. Rowley, P.D., Mehnert, H.D., Naeser, C.W., Snee, L.W., Cunningham, C.G., Steven, T.A., Anderson, J.J., Sable, E.G., Anderson, R.E., 1994. Isotopic ages and stratigraphy of Cenozoic rocks of the Marysvale volcanic field and adjacent areas, west-central Utah. U.S. Geol. Surv. Bull. 2071, 35 p. Rye, R.O., 1993. The evolution of magmatic fluids in the epithermal environment: the stable isotope perspective. Econ. Geol. 88, 733 – 753. Rye, R.O., 2005. A review of the stable-isotope geochemistry of sulfate minerals in selected igneous environments and related hydrothermal systems. Chem. Geol. 215, 5 –36 (this volume).

184

G.P. Landis, R.O. Rye / Chemical Geology 215 (2005) 155–184

Rye, R.O., Bethke, P.M., Wasserman, M.D., 1992. The stable isotope geochemistry of acid sulfate alteration. Econ. Geol. 87, 225 – 262. Sarda, P., Staudacher, T.J., Alle´gre, C.J., 1988. Neon isotopes in submarine basalts. Earth Planet. Sci. Lett. 91, 73 – 88. Sato, M., 1978. Oxygen fugacity of basaltic magmas and the role of gas-forming elements. Geophys. Res. Lett. 5, 447 – 449. Sato, M., Wright, T.L., 1966. Oxygen fugacities directly measured in magmatic gases. Science 153, 1103 – 1105. Shibata, T., Takahashi, E., Matsuda, J., 1998. Solubility of neon, argon, krypton, and xenon in binary and ternary silicate systems: a new view on noble gas solubility. Geochim. Cosmochim. Acta 62, 1241 – 1253. Smith, A., 1941. The geochemistry and paragenesis of the ores of the Cactus mine, Kern County, California. PhD thesis, California Inst. Technol., Pasadena. Smith, R.B., Bruhn, R.L., 1984. Interplate extensional tectonics of the eastern Basin-Range: inferences on structural style from seismic reflection data, regional tectonics, and thermal mechanical models of brittle–ductile deformation. J. Geophys. Res. 89, 5733 – 5762. Snyder, G., Poreda, R., Hunt, A., Fehn, U., 2001. Regional variations in volatile composition: isotopic evidence for carbonate recycling in the Central American volcanic arc. Geochem. Geophys. Geosyst. (G3) 2, 1 – 25. (2001GC000163). Stoffregen, R., 1987. Genesis of acid-sulfate alteration and Au–Cu– Ag mineralization at Summitville, Colorado. Econ. Geol. 82, 1575 – 1591. Stoffregen, R.E., Alpers, C.N., Jambor, J.L., 2000. Alunite–jarosite crystallography, thermodynamics, and geochronology. In: Alpers, C.N., Jambor, J.L., Nordstrom, D.K. (Eds.), Sulfate Minerals—Crystallography, Geochemistry, and Environmental Significance, Rev. Mineral. Geochem., vol. 40, pp. 453 – 479.

Symonds, R.B., Reed, M.H., 1993. Calculation of multicomponent equilibria in gaseous–solid–liquid systems: calculation methods, thermochemical data, and applications to studies of hightemperature volcanic gases with examples from Mt. St. Helens. Am. J. Sci. 293, 758 – 864. Symonds, R.B., Rose, W.I., Gerlach, T.M., Briggs, P.H., Harmon, R.S., 1990. Evaluation of gases, condensates, and SO2 emissions from Augustine volcano, Alaska: the degassing of a Cl-rich volcanic system. Bull. Volcanol. 52, 355 – 374. Symonds, R.B., Reed, M.H., Rose, W.I., 1992. Origin, speciation, and fluxes of trace-element gases at Augustine volcano, Alaska: insights into magma degassing and fumarolic processes. Geochim. Consmochim. Acta 56, 633 – 657. Symonds, R.B., Rose, W.I., Bluth, G.J.S., Gerlach, T.M., 1994. Volcanic-gas studies: methods, results, and applications. In: Carroll, M.R., Holloway, J.R. (Eds.), Volatiles in Magmas, Rev. Mineral., vol. 30, pp. 1 – 66. Symonds, R.B., Gerlach, T.M., Reed, M.H., 2001. Magmatic gas scrubbing: implications for volcano monitoring. J. Volcanol. Geotherm. Res. 108, 303 – 341. Troxel, B.W., Morton, P.K., 1962. Mines and mineral resources of Kern County, California. Calif. Div. Mines Geol. County Rept. 1, 370 p. Weiss, R.F., 1970. The solubility of nitrogen, oxygen, and argon in water and seawater. Deep-Sea Res. 17, 721 – 735. Weiss, R.F., 1971. Solubility of helium and neon in water and seawater. J. Chem. Eng. Data 16, 235 – 241. Yatsevich, I., Honda, A., 1997. Production of nucleogenic neon in the Earth from natural radioactive decay. J. Geophys. Res. (Solid Earth) 102 (B5), 10291 – 10298. Zimbelman, D.R., Rye, R.O., Landis, G.P., 2000. Fumaroles in ice caves on the summit of Mount Rainier—preliminary stable isotope, gas, and chemical studies. J. Volcanol. Geotherm. Res. 97, 457 – 473.